WEBVTT

00:00:00.000 --> 00:00:03.040
When people say they had bad workflows or bad outputs,

00:00:03.200 --> 00:00:10.080
it's because we're relying on the model to be like us. It's not. We need to be very, very, very, very, very explicit.

00:00:10.080 --> 00:00:20.935
We're not going to allow the model to be creative. And with these agents, there's what looks cool and there's what works. I used to have a bookkeeper. I don't have a bookkeeper anymore because I literally dumped

00:00:20.935 --> 00:00:24.295
all my statements. And then I asked them, look at the transactions

00:00:24.295 --> 00:00:34.990
and look at what sounds like a restaurant or a food spot, and then remove that. Let's say that someone's watching this and they want to build their first AI agent that's actually productive.

00:00:35.070 --> 00:00:36.670
What in your mind

00:00:36.910 --> 00:00:41.070
is the step one? Awesome. You know what? I'm just gonna open Codex,

00:00:41.390 --> 00:00:43.470
and we're going to go through it together.

00:00:43.790 --> 00:00:52.275
Okay. So a lot of you will be watching and seeing this show for the first time, and so I wanted to ask for a quick favor, and that's to subscribe to the channel.

00:00:52.435 --> 00:01:01.395
And if you do, the promise that I'll make is we're gonna take everything that we do here to the next level. That means scale the guest, scale the storytelling,

00:01:01.475 --> 00:01:02.115
the production,

00:01:02.790 --> 00:01:04.870
every little detail

00:01:05.030 --> 00:01:10.390
of what makes this show great so that we bring you an even better experience.

00:01:10.710 --> 00:01:12.550
You hit that subscribe button,

00:01:12.550 --> 00:01:16.230
we make this show better for you. Do we have a deal?

00:01:16.630 --> 00:01:20.315
Let's get back to the show. Here's where I wanted to begin.

00:01:21.355 --> 00:01:24.235
Because you and I, we had a conversation before this,

00:01:24.635 --> 00:01:29.755
and I think for most people that I speak to that I've tried using AI agents,

00:01:30.075 --> 00:01:32.315
it's not actually making them more productive.

00:01:33.140 --> 00:01:38.100
Like, it's not saving them time. It's not helping them make money or increase their output.

00:01:38.820 --> 00:01:43.060
And so if someone is listening with that feeling

00:01:43.060 --> 00:01:44.740
and with that experience

00:01:44.900 --> 00:01:46.340
of using AI agents,

00:01:47.175 --> 00:01:51.735
what is gonna be the value to them of listening to this conversation to the end?

00:01:52.455 --> 00:01:55.095
Yeah. I think there's a lot of, like, um,

00:01:55.255 --> 00:01:58.055
bad information out there on how to use these tools.

00:01:58.295 --> 00:02:04.000
Um, a lot of times, people are scaling for what looks cool, but really what you want is to scale for productivity.

00:02:04.000 --> 00:02:37.220
And just like in anything in life, simplicity always wins. And with these agents, when you use them with a simplistic mindset, when you have an end goal in mind, and we're gonna, you know, discuss that and architect that together, you start to realize, oh, this is actually a productive tool. Right? A lot of people are attaching 15 different workflows from other people, not learning how to build their own. So if there's one thing I want people to get from this podcast is they get an agent, and they know how to build a workflow that works for them. It might not work for me and you, but it's gonna work for them. And at the end of the day, that's what matters. My workflow is mine, and yours is yours.

00:02:37.700 --> 00:02:39.780
You know what? What you just mentioned,

00:02:39.940 --> 00:02:41.540
and I know that you and I spoke about this,

00:02:42.275 --> 00:02:44.275
it's actually, like, a contrarian

00:02:44.275 --> 00:02:44.915
take

00:02:45.235 --> 00:02:48.035
because most videos creators

00:02:48.035 --> 00:02:52.915
content that we see is people that are using, like, 50 different agents

00:02:53.155 --> 00:02:57.955
that the architecture and the system is incredibly complicated and complex.

00:02:59.420 --> 00:03:05.180
Why is it that you have this different perspective? Like, why is it that simplicity actually wins here?

00:03:05.660 --> 00:03:10.220
Yeah. So my mom always told me to be nice, so I'm gonna be nice to people who have contrarian takes.

00:03:11.095 --> 00:03:43.525
It's either one or two things. Um, complexity looks cool. Right? Like, when you have, like, 15 different sub agents, I'm sure, like and I'll do it too from time to time. I'll feel like Iron Man, like Tony Stark. Right? But when it comes to actually being productive, there's what looks cool and there's what works. Right? And what works oftentimes is boring. I'm sure you would say in business with all the people you've interviewed in yourself, sometimes it's the boring stuff that gets to the finish line. It's literally no different with using these tools. It's a matter of a lot of people, especially on YouTube, and I'm a YouTuber. Right? Like, I do content.

00:03:43.685 --> 00:03:55.660
The extravagant stuff gets clicks, and it's cool, and it's and it's fun, and it, you know, sort of paints this picture of a future that we can maybe have thousands of agents doing, like, you know, thousands of things, and you're just home sipping a pina colada.

00:03:55.660 --> 00:04:02.940
Uh, but the truth of the matter is when you want to use these agents, there's a couple things that are misunderstood.

00:04:02.940 --> 00:04:04.380
A, they're very powerful.

00:04:04.780 --> 00:04:07.085
B, they're very smart. C,

00:04:07.165 --> 00:04:08.365
this sounds contradictory,

00:04:08.365 --> 00:04:13.325
they're also very dumb, and, d, your influence matters a lot more than you think. Right?

00:04:13.565 --> 00:04:31.060
When you take all this into consideration, 15 sub agents don't make sense because you're not doing 15 things at the same time. Right? So I hope, like, in this pod, we sort of have, like, a mindset shift. People give it a week or two to test this out, and they come back to this video and be like, you know what? The guy with the messed up here was right.

00:04:31.860 --> 00:04:35.540
I love it. You know what? Because you you talk about the the mindset shift,

00:04:35.885 --> 00:04:37.565
and I think it's important.

00:04:38.125 --> 00:04:40.285
Can you just make it clear for people?

00:04:40.765 --> 00:04:48.285
If they're able to go through that mindset shift and some of the things that you and I speak about over the next eighty or ninety minutes, they're able to implement,

00:04:49.430 --> 00:04:52.070
what is gonna be the tangible impact

00:04:52.070 --> 00:04:56.390
in their lives? Like, what is it that they're gonna feel once this is implemented?

00:04:57.030 --> 00:05:18.290
Yeah. So, like, the first thing is you're actually going to see productivity with these tools, right, depending on, like, your personal work and stuff like that. Like, if you're a super busy person with a business and you have fifteen, twenty adamant things, then you're gonna love this. If you're someone who's just getting started out, you might see some value in some of the things. But the one thing I I look at and I think of is, like,

00:05:18.530 --> 00:06:03.985
investing in in in knowledge of these tools and how they work and sort of understanding high level how models and agents and stuff work feels like an investment for the future. Right? So, like, we can imagine next year from now based on the last year's growth, the models are gonna get better, the tools are gonna get better, agents are gonna be all over the place. It it it seems like anyone who's made an investment now in learning these things will be well equipped for the future. Right? So it's not just a now thing in my opinion. It's also sort of prepping for the future. Because I remember last year messing with these tools. A lot of them were garbage. Right? Like, they were so bad. But, you know, we kept like, people like me kept tinkering because you saw, like, okay. At some point, this thing's gonna get good, and now it's really good. So imagine where will we be six months, a year from now. So we're gonna experience,

00:06:04.145 --> 00:06:22.090
um, how to craft specific workflows and skills and all these thing things, and I'll explain all of that. But we're also sort of future proofing our mind so that when the latest and greatest tool comes, there's parallels we can draw from the previous tools that we've used. Yeah. You know, I think I think it's so good, and and you make that point about super proofing

00:06:22.250 --> 00:06:23.130
your mind

00:06:23.370 --> 00:06:24.970
and, like, being ready for the future.

00:06:25.795 --> 00:06:31.555
And so one thing, because you mentioned this when we spoke earlier, you said, I happen to be in a position

00:06:31.795 --> 00:06:36.195
that gives me unique insight on all of the stuff that's happening with AI

00:06:36.355 --> 00:06:39.075
and the software changes that we're experiencing.

00:06:39.810 --> 00:06:43.970
When you say that, I happen to be in a position that gives me a unique insight.

00:06:44.450 --> 00:06:46.610
Can you describe what that means?

00:06:46.930 --> 00:07:18.310
Or can you explain what that means? Yeah. So there's there's two things. First and foremost, you're not gonna get a biased take from me because none of these companies pay me. And the job I'm in, uh, I work for a back end as a service company that a lot of AI applications integrate to. Um, I see a lot of the progress made. I I, like, I see a lot of the updates made. I'm also very much a person who doesn't go outside. I go outside on Sundays for church. Other than that, like, my free time is spent reading the latest papers, seeing the latest updates, trying the latest tools. Right? So with, uh, when you take mere obsession

00:07:18.390 --> 00:07:26.155
plus my day job plus my content plus me, uh, not being bought by any of these companies, again, you know, pray for me. Hope hopefully, it happens.

00:07:27.275 --> 00:07:42.110
Yeah. I I I I have a unique perspective that's not biased, and, you know, I'll call garbage what's garbage, and I'll call great what's great. Yeah. You know what, Mike? Because you make you make the point that you're deeply inside this world. Like, you spend all of your time doing this.

00:07:42.510 --> 00:07:48.510
I think also the thing that I respect from you and why we I invited you on the show, like, I wanted you to do an episode with us,

00:07:48.910 --> 00:07:53.625
is you don't gatekeep information. You know, like, you you share information

00:07:53.945 --> 00:07:58.905
to people that are just starting out or even if they're more into this sort of thing.

00:07:59.865 --> 00:08:01.705
Can you just share with people

00:08:02.745 --> 00:08:04.665
why are you so open

00:08:05.420 --> 00:08:24.495
about sharing the information that you do? Because you're you're directly benefiting from using all of these tools, like, in your life. You don't necessarily have to share it. So so why are you so open in that? I remember getting into the space, Like, I felt really dumb. Like, I felt like, oh, man. All these people are so smart. They're so intellectual.

00:08:24.655 --> 00:08:42.250
Like like, they use all these big words. Like, they they construct these sentences that I have no understand no understanding in terms of what it means. And I remember, like, especially, like, early, like, 2021, 2022, I would watch videos by really, really smart people. We're talking about, like, developers, researchers, all that stuff, and I would force myself

00:08:42.410 --> 00:08:51.105
to listen to them. And this is pre chat, GPT. Then I would Google what they said trying to understand what it is they meant. So I went through a very painful process of learning,

00:08:51.505 --> 00:08:56.945
and I realized, oh, a lot of this stuff isn't really that hard. It just sounds hard.

00:08:57.265 --> 00:08:57.905
Um,

00:08:58.145 --> 00:09:12.100
and I I shared the story with you. Part of how I started the YouTube channel was I I was looking for a developer job, and, like, it just felt like no one was hiring or my resume wasn't good enough. So I I was like, okay. Maybe if I record a video of me explaining really deep technical concepts,

00:09:12.420 --> 00:09:28.805
um, that the recruiter watches and they see and they understand, then maybe they'll be like, okay. This guy's smart, they'll push me, like, to the forefront. That never happened. But what happened was that YouTube channel started to pick up. Um, and then I saw that, oh, people actually wanna learn about this stuff. Right? Like, a lot of people find technology

00:09:28.805 --> 00:09:29.525
fun,

00:09:29.685 --> 00:09:35.260
but it just feels like it it's above them, they don't understand, and they don't wanna make the investment in understanding.

00:09:35.580 --> 00:09:39.820
And, yeah, I just kinda made it like a thing where I was like, you know, I'm

00:09:39.820 --> 00:09:45.820
gonna explain it to one of the homies. Right? Like, I record my videos as if, uh, my bro or one of the homies from church was watching,

00:09:46.385 --> 00:09:54.385
And I guess that's helped a lot of people, so I'm very thankful for that. Yeah. Yeah. That's that's awesome. And I I I love the intention. Like, you can feel it in your work.

00:09:55.585 --> 00:09:58.385
You know what, Mike? I just wanna I wanna get straight into it.

00:09:59.110 --> 00:09:59.750
And

00:10:00.230 --> 00:10:03.190
in particular, because we started it off in this conversation,

00:10:03.350 --> 00:10:05.750
and this has been to be transparent,

00:10:05.830 --> 00:10:07.430
this has been my experience.

00:10:07.670 --> 00:10:12.150
Because a few weeks ago, we record an episode with our mutual friend, Riley Brown,

00:10:12.655 --> 00:10:22.815
and he talks about OpenClaw and how he's using it in his business and how how he has all these agents. And so I get super excited. I don't actually start with using OpenClaw.

00:10:22.895 --> 00:10:24.815
I start with Claude Cowork,

00:10:25.080 --> 00:10:32.440
and I spend a few hours setting up this workflow that's gonna help me do content or in my business with Claude Cowork.

00:10:32.680 --> 00:10:35.000
I set it up. It gives me an output.

00:10:35.240 --> 00:10:37.880
So I get to that point where it's given me an output.

00:10:38.200 --> 00:10:39.800
The output is mediocre

00:10:39.800 --> 00:10:40.360
at best.

00:10:41.405 --> 00:10:48.605
From that point, basically, from four weeks ago, I haven't used it again. And I think that that's like a common experience.

00:10:49.085 --> 00:10:49.885
And so

00:10:50.125 --> 00:10:51.645
for people like me,

00:10:53.100 --> 00:10:55.820
what is the thing that we're like misunderstanding

00:10:55.900 --> 00:11:03.340
or doing wrong when it comes to building these AI agents where we're not actually getting a useful output at the end of this long process?

00:11:04.060 --> 00:11:13.135
Okay. Real quick. If you're watching this video, I'm sure at some point you've had to create a presentation in your life. And if you're a non designer like myself,

00:11:13.375 --> 00:11:15.775
then that probably was not a good experience.

00:11:15.935 --> 00:11:19.135
The funny story is I actually started my career in consulting

00:11:19.580 --> 00:11:22.620
and I would only work in building presentations

00:11:22.620 --> 00:11:26.380
for clients, and it'd be this horrible manual experience.

00:11:26.380 --> 00:11:31.260
It would be so frustrating trying to get everything to look just right on the slides.

00:11:31.580 --> 00:11:33.580
But now with tools like Gamma,

00:11:34.005 --> 00:11:37.125
you simply type in a text prompt

00:11:37.365 --> 00:11:40.165
and it will design a beautiful presentation

00:11:40.165 --> 00:11:43.845
in exactly the way that you want it. And so as an example,

00:11:44.085 --> 00:11:48.085
recently I wanted to create a short form content strategy presentation

00:11:48.600 --> 00:11:50.200
to share with my team.

00:11:50.440 --> 00:11:54.200
And so I simply went into Gamma. I selected one of their templates,

00:11:54.440 --> 00:12:02.520
and I described how I wanted this presentation to look. And so what I asked Gamma to do was to build me a short form content presentation

00:12:03.205 --> 00:12:13.445
using some of the strategies from the top podcasts in the world. And you can see here what it came up with, and I actually think it's pretty impressive. You can see we have the platform strategy,

00:12:13.605 --> 00:12:18.490
even created this checklist for choosing viral clips, and you can see it looks great.

00:12:18.650 --> 00:12:22.970
And so if you wanna create good professional looking presentations

00:12:23.290 --> 00:12:30.965
without all of the manual design work that typically comes with it, then go to the link in the description and check out Gamma.

00:12:31.045 --> 00:12:33.445
That's Gamma using the link in the description

00:12:33.605 --> 00:12:36.005
to get started with them today.

00:12:36.165 --> 00:12:45.590
Thank you to Gamma for sponsoring this episode. Let's get back to the show. Yeah. I mean, I could share my screen and, like, diagram this out. Let's do it. It really comes from, um,

00:12:46.070 --> 00:12:49.190
like, when you understand how models work,

00:12:49.430 --> 00:12:58.790
um, it will all make sense. Right? So you can think of this giant blue box right here as a model. You can it could be Opus, uh, 4.7,

00:12:58.790 --> 00:12:59.590
which is Anthropix

00:13:00.045 --> 00:13:02.845
latest model, or it can be GPT 5.5,

00:13:02.845 --> 00:13:03.965
which is OpenAI's,

00:13:03.965 --> 00:13:07.645
uh, latest model. Both are really, really capable and smart models.

00:13:08.125 --> 00:13:09.565
The issue

00:13:09.805 --> 00:13:13.725
with a lot of people and how they use it is a lot of people

00:13:14.390 --> 00:13:29.430
think the model is similar to them in a sense where, you know, the model kinda sounds like us, acts like us. You might hear these stories where, like, oh, they did some testing and the model they got the model to blackmail one of the users and all that stuff. It it feels very human like,

00:13:30.195 --> 00:13:31.155
but it's not.

00:13:31.555 --> 00:13:38.675
And understanding this is important because when you understand where it lacks and where it's great at, it becomes a great copilot.

00:13:38.675 --> 00:13:44.355
Right? I think agents are a great copilot. You're still the pilot. You're still in charge, uh, but they're a great toolkit.

00:13:44.860 --> 00:13:49.660
Now one of the interesting things about how models work is they actually,

00:13:49.980 --> 00:13:52.940
um, don't think the way you and I think. Models,

00:13:53.020 --> 00:14:02.325
uh, predict tokens. Now a lot of people are gonna hear this and be like, oh, this nerd is about to speak some jibber jabber and it's gonna confuse me. I I I promise all this will make sense.

00:14:02.645 --> 00:14:04.325
Tokens are basically

00:14:04.725 --> 00:14:05.365
numbers,

00:14:05.925 --> 00:14:06.805
like weird looking

00:14:07.525 --> 00:14:27.660
and these numbers are plotted on a graph. If anybody remembers in grade ten, eleven, or nine, whatever, like, year it was, like, at math class, we'd have to draw these graphs, and then we have to plot them, and it was annoying. And I was like, why is this ever useful? It so happens that the researchers who build these models needed that stuff. So glad to hear that at least some of those things were useful.

00:14:28.505 --> 00:14:31.545
So what the model does, it it understands

00:14:31.545 --> 00:14:34.905
and predicts tokens. What does that mean? Let's say,

00:14:35.625 --> 00:14:38.825
Callum, I prompt what is the capital

00:14:38.825 --> 00:14:40.025
of France.

00:14:40.185 --> 00:14:53.810
Right? This is my prompt. What is the capital of France? You and I know that the capital of France is Paris. It's been taught to us. We have it stored in our brain, like, in in some part of the memory layer of our brain, and we know this instantly.

00:14:54.445 --> 00:15:12.010
The way model works a model works is it's going to when I type in this text, it's going to convert this text to a bunch of numbers like this. Again, how it converts? It doesn't even matter. I don't even know. But it's gonna convert these number these texts to the numbers. It's gonna plot the number on the graph.

00:15:12.490 --> 00:15:14.330
Mathematically, and this is impressive,

00:15:14.650 --> 00:15:16.730
the closest answer,

00:15:16.890 --> 00:15:18.010
which is Paris,

00:15:18.170 --> 00:15:30.675
happens to be a number very close on the graph. So that's what the model does. It doesn't think it doesn't, like, ponder the way you and I do. It literally maps words on a graph and finds the nearest one.

00:15:31.155 --> 00:15:36.450
I'm going someplace with this. It's going to map it, and then it's gonna be like, oh, the answer is Paris.

00:15:36.930 --> 00:15:44.530
There's even another test right now, um, especially with the Anthropic model. You can ask it. Um, I want to wash my car.

00:15:44.850 --> 00:15:47.405
There's a car wash 50 miles away.

00:15:47.805 --> 00:15:53.405
It's faster for me to walk than drive. What should I do? The model will tell you to walk.

00:15:53.885 --> 00:16:05.810
I'm going to car wash to wash my car, but the model will tell you to walk. So the the reason why I say this is the model does not think the way you and I think. So when you understand this, you have to understand

00:16:05.970 --> 00:16:27.315
the prompts that you give it, the text that you enter, the messages, the workflow, whatever it is that you construct has to be good. It it like, the quality of input has to be good because when the quality of input is good, the quality of output will be good. But if we talk to the model the way you and I talk to each other where there might be sarcasm,

00:16:27.395 --> 00:16:30.195
there might be context of understanding based on a relationship,

00:16:30.560 --> 00:16:38.560
You know, I can say something to you, and based on the context of our our relationship, it can mean totally something different than if I say it to somebody else.

00:16:38.880 --> 00:16:50.955
In the same way, the model is not human at all. It sounds human. It acts human. It's been trained on so much data where you think it's conscious, but it's not. So when you understand this, you understand you have a really smart,

00:16:51.195 --> 00:16:53.115
but at the same time dumb machine.

00:16:53.355 --> 00:16:56.715
So when people say they had bad workflows or bad outputs,

00:16:56.875 --> 00:17:08.210
it's because we're relying on the model to be like us. It's not. At its current state, it magnifies the abilities, the wisdom, and the knowledge that you have. So if you're someone who says, I I don't need to read books anymore,

00:17:08.450 --> 00:17:11.970
ta da. You gotta keep reading more and kinda gotta keep growing more knowledge.

00:17:12.050 --> 00:17:36.310
But this is how models work. So when you understand this, it's going to help, like, especially when we start to craft a workflow. It's going to help us craft a workflow because we're gonna understand we need to be very, very, very, very, very explicit. We're not going to allow the model to be creative. We're going to give it our creative sauce, and we're going to want it to replicate that. Just so just one thing, Mike, before you before you go on, because I think this is so good and important.

00:17:36.790 --> 00:17:41.750
When you say, um, it really comes down to, like, the the prompts and, like, the inputs.

00:17:42.125 --> 00:17:45.245
Right? It's like good inputs versus bad inputs.

00:17:45.405 --> 00:17:47.805
Mhmm. Can you crystallize for people exactly

00:17:47.885 --> 00:17:51.085
like, what is the distinction between what makes a good input

00:17:51.165 --> 00:17:55.885
or how someone can know if they're giving the AI a good input versus a bad out input?

00:17:56.710 --> 00:18:00.310
Yeah. I I could explain that. So let's say we have this line,

00:18:00.470 --> 00:18:02.790
uh, where on this spectrum,

00:18:02.950 --> 00:18:05.590
this is a person. They have no agency.

00:18:05.910 --> 00:18:08.310
Right? And on this end of the spectrum,

00:18:08.550 --> 00:18:10.055
they have very

00:18:10.055 --> 00:18:13.255
much agency. I know that just broke English. I apologize.

00:18:13.255 --> 00:18:14.055
Um,

00:18:14.135 --> 00:18:18.375
but, basically, let's say this individual like, we're looking to hire somebody.

00:18:18.695 --> 00:18:28.240
And as a business, we obviously want someone who has, like, you know, a lot of agency. And to have agency means you're given a problem. You might not have all the context,

00:18:28.480 --> 00:18:41.455
uh, but you go out of your way to get the context. Um, you have this feedback loop where you keep on iterating on this feedback loop, and at some point, you're going to achieve success. Right? Like, for example, if I hired an assistant,

00:18:41.615 --> 00:18:45.935
uh, one thing that I may do, and I'm gonna show us how to do, is generate

00:18:46.335 --> 00:18:47.775
a a report

00:18:47.935 --> 00:18:49.375
for sponsored

00:18:49.455 --> 00:18:57.420
videos. Now I'll give context on this. I run a YouTube channel, and one thing I use AI for is after a video's gone live, especially if it's a sponsored video,

00:18:57.660 --> 00:19:03.180
I'm going to send out a report of how the video performed. Right? Like, sponsor

00:19:03.340 --> 00:19:09.325
wants to see if they got a good investment on their money. Now if I told this to a very high agency person,

00:19:09.725 --> 00:19:46.655
they're probably gonna research, okay. What does reports in the YouTube space look like? Right? What things do I need to report? Is it just views, or is it link clicks? Do I calculate a specific thing? Right? Someone who has high agency will go through the motions, will probably continue to iterate on a specific thing and generate a report that might be decently good. Someone with low no agency will probably be like, yeah. Like, here's the video. Here's the link clicks. Done. Right? So, um, we we we see this as humans. Right? Like, you you wanna work with, like, the very high agency people and the, you know, the ones who, like, make the right decisions, ask the right questions.

00:19:47.580 --> 00:19:48.460
Models

00:19:48.460 --> 00:19:49.740
are somewhere here.

00:19:51.340 --> 00:19:59.900
Right? And the reason why I say this is it is going to like, it it will produce the amount of effort you give it in the question you ask.

00:20:00.545 --> 00:20:04.705
Right? So if I say generate a report for a sponsored video,

00:20:05.105 --> 00:20:08.225
it's going to generate it as if someone with no agency did.

00:20:08.865 --> 00:20:09.505
But

00:20:09.905 --> 00:20:12.785
if I, let's say, change the prompt to

00:20:13.185 --> 00:20:14.865
generate a report

00:20:15.105 --> 00:20:15.505
for

00:20:15.960 --> 00:20:16.840
a video.

00:20:17.560 --> 00:20:21.560
Right? And then I add let me just make this a long line. I add,

00:20:21.720 --> 00:20:22.360
um,

00:20:22.520 --> 00:20:25.000
let me add a new line. First, get

00:20:25.480 --> 00:20:26.920
YouTube views,

00:20:28.360 --> 00:20:30.280
then get

00:20:29.785 --> 00:20:31.065
link clicks,

00:20:31.865 --> 00:20:33.545
then calculate

00:20:33.945 --> 00:20:34.985
CTR,

00:20:35.305 --> 00:20:36.505
then generate

00:20:36.745 --> 00:20:37.545
HTML

00:20:37.545 --> 00:20:39.785
page for me to view

00:20:40.185 --> 00:20:41.305
the stats.

00:20:41.945 --> 00:20:42.585
So

00:20:42.825 --> 00:20:43.865
this

00:20:44.200 --> 00:20:45.160
explanation

00:20:45.320 --> 00:20:48.280
and this get generate a report for a video,

00:20:49.480 --> 00:20:54.200
to me and you, especially if maybe we've worked together, you know, we're both creators,

00:20:54.360 --> 00:20:57.560
generate a report for a video might make sense to us.

00:20:58.815 --> 00:21:02.335
But to someone who has no agency, you're gonna have to give them step by step direction.

00:21:02.975 --> 00:21:09.695
The model is pretty much the same way. You're going to have to be very like, you have to think of this as a very junior,

00:21:09.935 --> 00:21:15.400
no agency person, but that is very capable and has all the knowledge in the world consumed

00:21:15.400 --> 00:21:16.440
in their brain.

00:21:17.080 --> 00:21:32.765
So this has to, uh, the the the the way you prompt has to shift, you can't give it one liners. You can't, uh, say just do this for me. It doesn't know you or understand you. Right? You're going to have to give it step by step by step instruction

00:21:32.765 --> 00:21:39.485
at least one time. And then I'm gonna show us later down in the in in in in the video how we can take this,

00:21:39.725 --> 00:21:41.565
convert this to what's called a skill.

00:21:42.290 --> 00:21:43.570
This is stored

00:21:43.650 --> 00:22:05.375
in the agent, and the agent can call this again and again when it needs to. But the very first time, like hiring a new employee, we're going to need to train them. And this is the big difference on, like, good inputs versus bad inputs. Yeah. You know what? You you just made the point, and I think it's actually the perfect way to look at it, and it's what made it click for me. 100%.

00:22:05.375 --> 00:22:08.095
100 per like, that's literally the perfect analogy.

00:22:08.735 --> 00:22:09.375
Yeah.

00:22:09.695 --> 00:22:12.470
Okay. Where do we go to next? So there's

00:22:12.710 --> 00:22:14.470
I would say there's a spectrum,

00:22:14.950 --> 00:22:17.190
um, when it comes to these agents,

00:22:17.430 --> 00:22:18.630
and the spectrum,

00:22:19.030 --> 00:22:20.710
um, really is

00:22:21.350 --> 00:22:23.910
diff the difficulty. Right? How difficult

00:22:24.255 --> 00:22:30.015
is it to set up? We'll actually just have one line here. How difficult is it to set up? If, like, this is super

00:22:30.015 --> 00:22:30.735
difficult

00:22:30.815 --> 00:22:33.615
and then this is, uh, easy,

00:22:34.655 --> 00:22:36.415
uh, the following agents

00:22:36.815 --> 00:22:37.695
fall as follows.

00:22:38.180 --> 00:22:39.060
OpenAI

00:22:39.060 --> 00:22:39.860
and,

00:22:40.100 --> 00:22:40.740
um,

00:22:40.980 --> 00:22:43.780
OpenAI Codex and, uh, Anthropic,

00:22:44.020 --> 00:22:45.140
their tool,

00:22:45.300 --> 00:22:47.380
I would say, is somewhere,

00:22:47.780 --> 00:22:57.385
like, not in the like, not on the easy spectrum, but somewhere, like, right here. Meaning, um, there is, like, a back and forth. Um, there is, like, a cost of time,

00:22:57.705 --> 00:23:02.265
um, that you're going to put. But I believe, like, especially after we go through an example,

00:23:02.425 --> 00:23:04.585
um, you're going to feel like this is super easy.

00:23:05.570 --> 00:23:06.850
These OpenClaws,

00:23:06.850 --> 00:23:09.010
Hermes, these, like, new agents,

00:23:09.250 --> 00:23:11.730
I would say, especially for someone nontechnical,

00:23:11.890 --> 00:23:21.555
are on the difficult side, uh, when it comes to I'll just say OpenClaw and Hermes, are on the difficult side when it comes to setting up. But, like, using them directly,

00:23:21.795 --> 00:23:28.675
um, the setup is a bit difficult. Now there's pros to that. The pros to that is I can have a really highly customizable

00:23:28.835 --> 00:23:31.555
agent, like, to the core, like, personality,

00:23:31.555 --> 00:23:37.940
how it works. Like, everything is fully customizable, and I own, like, the code. With open ananthropic,

00:23:37.940 --> 00:23:41.300
you're sort of tied to an ecosystem. You're tied to specific model.

00:23:41.460 --> 00:23:43.540
Um, so these are the pros and cons.

00:23:43.940 --> 00:23:44.980
For

00:23:44.755 --> 00:23:51.635
first timers or people who are not technical or people who are giving this, like, a try and you're pretty early on, I

00:23:52.675 --> 00:24:00.275
before we chatted, I was like, oh, coworker seems to be the one, but I've changed my mind. I think Codex OpenAI's

00:24:01.930 --> 00:24:02.970
OpenAI's

00:24:02.970 --> 00:24:03.850
Codex

00:24:03.930 --> 00:24:13.370
is the best place to start. A, if you have an itch to build an application, a web or a mobile application, it's a great place to do so. But b, they've started to really focus on

00:24:13.995 --> 00:24:21.435
this concept of being like a super app where you can use it for knowledge work. Uh, you can use it for workflows and tasks. You can connect it to external

00:24:21.435 --> 00:24:21.995
tools.

00:24:22.715 --> 00:24:40.510
And that way, you get the both of the best of both worlds where I can maybe develop apps one day if I want to, or I can also automate my life and my business and my personal if I want to. Right? So, um, for today's video and probably for the near future, um, I would say Codex is the best bang for your buck. So so so, Mike, before we get into it, because this is interesting,

00:24:41.215 --> 00:24:45.295
you and I spoke, I wanna say, three days ago, two days ago.

00:24:45.935 --> 00:24:47.215
I wanna understand

00:24:47.535 --> 00:24:51.375
and and and when you and I spoke initially, you said that, um,

00:24:51.935 --> 00:25:01.800
your belief was that for beginners looking to start out, the easiest tool and the simplest way to get started if you wanna use AI agents is Claude Cowork.

00:25:02.040 --> 00:25:05.080
And so I wanna understand, what is it that has happened

00:25:05.640 --> 00:25:10.040
in these last three days that your thinking has shifted? Because I know Codex,

00:25:11.315 --> 00:25:15.875
the developments that they've made to are actually pretty new. Like, it's getting better all the time.

00:25:16.355 --> 00:25:19.075
What has happened that your position has shifted?

00:25:19.955 --> 00:25:24.835
So the biggest one the biggest one is the subsidy of credits.

00:25:26.560 --> 00:25:38.560
What that means is, like, when you subscribe to the o to the Codex $100 a month plan or $200 a month plan or $20 a month plan, and same thing for Anthropic, you're not actually getting $200

00:25:38.560 --> 00:25:42.115
of value. You're getting, like, 3 to $5,000

00:25:42.115 --> 00:25:45.875
worth of compute. For example, the Claude, uh, $200 subscription,

00:25:46.115 --> 00:25:48.275
you are technically getting $5,000

00:25:48.275 --> 00:25:55.875
worth of compute. Meaning, they are viscerally burning money. Um, you are getting, like, the best value on the $200 month subscription.

00:25:56.650 --> 00:26:01.690
Recently, the last couple of days and especially, like, the last two days, um, Anthropic

00:26:01.690 --> 00:26:05.290
has been, like, cutting down on this. Right? And

00:26:05.530 --> 00:26:11.450
OpenAI, on the other hand, has been increasing rate limits. Right? So when I look at this from a cost perspective,

00:26:11.985 --> 00:26:23.585
if both of them are equally just as good, what I want is I want to use the platform that's going to give me the most compute. Meaning, I can keep going back and forth with it, and I don't get rate limited. I don't have to wait till next week to use it again.

00:26:24.160 --> 00:26:25.280
And OpenAI's

00:26:25.280 --> 00:26:28.160
rate limit, Codex's rate limit, is extremely,

00:26:28.160 --> 00:26:36.080
extremely generous. Right? So because I have no loyalty to any of these applications because none of them pay me, I'm going to pick the best one for my pockets.

00:26:36.320 --> 00:26:38.560
And that's when I was testing both of them yesterday.

00:26:39.105 --> 00:26:50.065
Codex's rate limits were just great. And to your point, they pushed a lot of updates recently. So it like, based on all that, it felt like the best choice to recommend. So so simply

00:26:50.145 --> 00:26:50.545
put,

00:26:52.020 --> 00:26:55.140
Codex at this point in time that you and I are speaking

00:26:55.620 --> 00:26:57.940
will allow someone to build more,

00:26:58.260 --> 00:27:18.315
to to go back and forth with it more, to build more things for the same or similar price than what Anthropic did with Claude Cowork or Claude Code. Yes. Exactly. You're going to get more value for the for the subscription you subscribe to with, uh, OpenAI. Before we get deeper into it and how you would go about building your first,

00:27:18.555 --> 00:27:22.075
like, AI agent using codex that's actually productive

00:27:22.155 --> 00:27:25.370
and useful, whether that's making money or saving you time.

00:27:25.850 --> 00:27:29.130
Can you just share some examples with people

00:27:29.930 --> 00:27:36.010
of what codex is actually capable of? Like, use cases that you've seen either for yourself

00:27:36.685 --> 00:27:40.125
or with peers, friends, creators,

00:27:40.365 --> 00:27:41.565
uh, people online

00:27:41.805 --> 00:27:44.685
of, like, what they've been able to build with Codex.

00:27:45.245 --> 00:27:47.885
For mine, there's a couple of things. First and foremost,

00:27:48.125 --> 00:27:51.005
the, uh, biggest, like, time sync for me,

00:27:51.325 --> 00:27:52.765
um, was

00:27:52.380 --> 00:27:56.300
going back and forth, uh, was not even going back and forth. It was researching

00:27:56.620 --> 00:28:08.435
sponsors. Right? There are a lot of people that will email me, especially in the space I'm in, uh, because, like, everyone's building an AI now. There's, like, thousands of emails I get per day. Like, my email, unfortunately, is has been vicinerated.

00:28:08.595 --> 00:28:11.315
So I have this one specific workflow

00:28:11.315 --> 00:28:12.115
where

00:28:12.195 --> 00:28:16.275
every single day, right, around, I would say, 10AM,

00:28:16.675 --> 00:28:19.075
uh, I have a workflow that does the following.

00:28:19.315 --> 00:28:20.915
Check my emails.

00:28:20.995 --> 00:28:21.315
Right?

00:28:22.520 --> 00:28:23.160
Check

00:28:23.400 --> 00:28:27.560
if any email so first, it's gonna check my emails, then it's gonna check if any email,

00:28:27.720 --> 00:28:30.040
um, is from a potential

00:28:30.040 --> 00:28:31.000
sponsor.

00:28:31.080 --> 00:28:39.565
Now the agent's able to catch this because usually they'll put it in the subject line or be in one of the emails. Like, it's pretty easy to catch the sponsor.

00:28:39.725 --> 00:28:48.380
I'm about to share some secret sauce. I probably shouldn't have, but I'm I'm gonna share some secret sauce. Then what my agent's going to do is it's going to research if the company's raised the money.

00:28:48.860 --> 00:28:50.060
I'm gonna show why.

00:28:50.700 --> 00:28:53.500
If the company has raised

00:28:53.580 --> 00:28:54.300
money.

00:28:54.460 --> 00:28:55.660
First and foremost,

00:28:55.900 --> 00:28:57.420
um, in

00:28:57.580 --> 00:29:00.540
this AI space, dev tool space, software space,

00:29:01.075 --> 00:29:03.155
especially for my niche,

00:29:03.475 --> 00:29:06.275
um, which is, like, technical and nontechnical

00:29:06.275 --> 00:29:08.595
people, more so technical people,

00:29:08.995 --> 00:29:15.000
um, that's a valuable audience. Right? And a lot of the companies that wanna sponsor a channel like mine raise

00:29:15.320 --> 00:29:24.040
are raising a lot of money now. And if, obviously, you raise the money, there's two things. First and foremost, you're a legit company. Right? So we've verified legitimacy

00:29:24.040 --> 00:29:35.135
here. Well, most likely, there's I'm sure there's scams that raise money, but the likelihood of it being a legit company is high, number one. And number two, they have money to spend, so the price is gonna be high. Uh,

00:29:35.375 --> 00:29:48.000
so I get my agent to do this, and then my agent has its own email. Right? I use a tool called agent mail, and I'll explain, like, different tools to integrate, uh, later in the video. My agent has its own email,

00:29:48.880 --> 00:29:51.040
and then it's going to email me,

00:29:51.600 --> 00:29:58.515
and this is a private email I have that no one has access to but my agent. It's going to email me if the sponsor

00:29:58.515 --> 00:29:59.555
is worth it.

00:29:59.955 --> 00:30:02.675
So what this does is this is basically checking

00:30:02.675 --> 00:30:05.075
if the email I got from a sponsor,

00:30:05.395 --> 00:30:09.155
is it worth it, or is it not? Like, is it worth looking into it? Is it worth responding?

00:30:09.780 --> 00:30:11.140
And I would say

00:30:11.540 --> 00:30:13.940
nine times out of 10, this works.

00:30:14.180 --> 00:30:20.740
Right? The first few times I had to really massage it. Right? There were a couple rules I added. For example,

00:30:21.140 --> 00:30:25.060
most likely, a sponsor that hits me up with an @gmail.com

00:30:25.735 --> 00:30:31.335
could be, like, a scam because there are scams out there. Like, there's a friend of mine who's a YouTuber who connected

00:30:31.415 --> 00:30:51.440
to a landing page that was like, oh, connect to your YouTube to so he can see your analytics, and, like, he got hacked. Right? So I have to be careful with these things. So something like this saves me a ton of time. Another one, and this is the one we're going to build, is the sponsor report. And the reason why I like these is because this is the type of workflow

00:30:51.520 --> 00:31:08.855
that everyone watching this can set up today. There is some sort there's data in your life that is scattered between multiple sources that can be aggregated together to be used a piece of information that is valuable to you. Right? Reports aren't just boring old documents. It's an aggregation of information

00:31:08.935 --> 00:31:14.740
that should be useful to you. And the report that I generate is you get YouTube views,

00:31:14.980 --> 00:31:16.580
right, for a specific video,

00:31:17.140 --> 00:31:18.420
views for a video.

00:31:19.780 --> 00:31:23.615
You're going to get link clicks for the sponsor.

00:31:24.495 --> 00:31:27.055
You're going to calculate

00:31:27.455 --> 00:31:28.415
all the

00:31:28.735 --> 00:31:30.575
nerdy marketing CTRs.

00:31:30.575 --> 00:31:35.295
I don't even know half what half that stuffs mean, uh, and other metrics.

00:31:36.190 --> 00:31:39.070
Generate a HTML page.

00:31:39.070 --> 00:31:40.430
Right? This,

00:31:40.590 --> 00:32:03.485
again, was something that I dreaded that I had to do. And it also got to a point, you know, praise the Lord, I started to grow and a lot of sponsors started to hit me up. And I thought, you know, I'm an agent. I don't get tired. Let me accept 15 of these. And now I'm overwhelmed. I have 15 different things to generate. This saved me a ton of time, made me super productive.

00:32:03.565 --> 00:32:30.625
And this is the best example I wanna go with because, again, I don't want people to watch this. If you're watching this, I don't want you to watch this and build the same thing that I built. What I want you to see is I I purposely picked an built an example, uh, picked an example that's going to show you how to think about it, and then I want you to dream. I want you to like, whatever it is in your life, you could build a workflow and to build that. Yeah. You know, it's such a good breakdown, and and a few things that just listening to you,

00:32:31.185 --> 00:32:41.340
use cases that came to mind. I'm even just thinking and and you confirm or deny. This has just came to my mind. Even if people are looking to kinda track their, uh, like, personal expenses

00:32:41.580 --> 00:32:43.820
in, like, a more systematic,

00:32:44.300 --> 00:32:48.905
uh, way. If you're looking to get into content creation, you post on Instagram

00:32:48.985 --> 00:32:53.385
and you wanna get, like, weekly, monthly, I don't know, quarterly updates

00:32:53.385 --> 00:32:57.065
on, like, your analytics, what's working versus what's not working.

00:32:57.385 --> 00:33:01.790
To your point, there's so many places in our life where we're getting

00:33:02.110 --> 00:33:03.870
data or information

00:33:04.030 --> 00:33:05.390
that we're not tracking.

00:33:05.790 --> 00:33:13.125
It could even be, like, meeting notes from, uh, like, your weekly meetings at work. You want it condensed into a brief.

00:33:13.365 --> 00:33:32.770
No. Even food that you're eating, like, you like, you said it perfectly. Like, you know how there are these expense tracker apps that you gotta take the picture and stuff? You can use the agent. You can literally take a bunch of pictures of your receipts throughout the day, dump it into the agent, and tell it, oh, put it in a spreadsheet, and you could download that spreadsheet. Like, anywhere you can ingest information,

00:33:33.330 --> 00:33:35.490
it can be turned into something valuable.

00:33:35.730 --> 00:33:38.530
And that's what I want people to think of when they think of workflows.

00:33:38.935 --> 00:33:52.055
Think of all this information that exists around you, uh, around your business and personal life. Ask yourself, are you efficient with the information? The answer is no. And if the answer is no, how can I feed this to an agent

00:33:52.390 --> 00:33:54.630
and produce something actionable

00:33:54.630 --> 00:34:06.790
that is actually beneficial? Whether it's condensed information or it's actual work done. Like, that's like if people can think about that, then, you know, this is a w episode. You know what? Mike, here's here's where I wanna go,

00:34:07.915 --> 00:34:08.715
which is

00:34:09.115 --> 00:34:14.315
let's say that someone's watching this. They have a few hours, you know, coming up this weekend

00:34:14.875 --> 00:34:16.635
or like a free evening,

00:34:17.035 --> 00:34:21.355
and they want to build their first AI agent that's actually productive.

00:34:21.600 --> 00:34:27.200
Like, either takes work off their plate or helps them do more in less time.

00:34:28.000 --> 00:34:34.080
What in your mind is the step one? Like, where should they begin if they're gonna use Codex?

00:34:34.765 --> 00:34:39.965
Yeah. So there's two things, and, like, I I want people to make an honest assessment of themselves.

00:34:40.525 --> 00:34:41.405
Step one,

00:34:41.645 --> 00:34:42.525
if you

00:34:43.005 --> 00:34:43.645
are

00:34:44.365 --> 00:34:47.965
like, point number point one point I wanna make, if you are delusional,

00:34:49.460 --> 00:34:51.380
column shirt shirt says, meaning,

00:34:52.100 --> 00:35:02.500
you you you just have this drive where you're going to figure like, you don't need to be motivated to do things. You're going to drop to step number two, which is we're gonna build a workflow.

00:35:03.125 --> 00:35:05.125
We're going to build a workflow.

00:35:05.365 --> 00:35:15.365
But if you're someone who needs a little bit of motivation, you need to watch a David Goggins video before you get into it. I'm actually going to tell you to open Codex

00:35:16.500 --> 00:35:18.580
and create a random app.

00:35:19.460 --> 00:35:23.060
Here's why I say this, and I know this has nothing to do with workflows.

00:35:23.380 --> 00:35:28.100
The reason why I say this is I found, uh, Calum, especially with people who are nontechnical,

00:35:28.500 --> 00:35:38.625
when they just create something random. It doesn't have to be polished. It doesn't have to be finished. It doesn't have to be live. But when they create something random and and they see it and, like, it's functional,

00:35:38.945 --> 00:35:51.050
the dopamine hit you get is the equivalent to whatever class a substance is popular now. Like, there is just this excitement and drive and, like like, I remember I I got one of my church friends to do this, and he

00:35:51.450 --> 00:35:53.130
hasn't stopped since.

00:35:53.210 --> 00:35:57.690
Right? So if you are someone who is like, man, I've tried these tools.

00:35:58.205 --> 00:36:00.045
I'm just I'm tired.

00:36:01.645 --> 00:36:14.890
You're gonna open Cortex. I'm gonna show people how to do that. Um, but for anyone else, um, who's just ready to go, there's a couple there's one specific tool we're gonna need. Now a lot of people don't share this tool, so I'm gonna share this tool.

00:36:15.130 --> 00:36:19.610
And this tool is called Composeo. You're gonna go to composeo.dev,

00:36:19.770 --> 00:36:24.090
and you're going to make an account. What Composeo does is Composeo

00:36:24.090 --> 00:36:24.890
becomes

00:36:25.130 --> 00:36:26.650
and

00:36:26.255 --> 00:36:29.775
this is gonna be a new term for people. It's gonna it's called a tool router.

00:36:30.175 --> 00:36:33.055
And, basically, what this becomes is this becomes

00:36:33.135 --> 00:36:37.615
like the layer that allows our AI agent to connect to anything.

00:36:37.775 --> 00:36:39.375
So let's say I'm using

00:36:40.090 --> 00:36:40.970
OpenAI.

00:36:40.970 --> 00:36:42.650
Right? We'll call it OAI,

00:36:42.810 --> 00:36:46.570
and I want OpenAI to connect to my analytics platform,

00:36:46.890 --> 00:36:53.930
um, called Dubb. Right? I use an analytics platform called Dubb. What Composio does is I give OpenAI instructions

00:36:53.930 --> 00:36:58.065
that Composio gives me. OpenAI connects to Composeo.

00:36:58.305 --> 00:36:59.265
Composeo

00:36:59.265 --> 00:37:00.945
then connects to Dubb,

00:37:01.025 --> 00:37:01.745
and then

00:37:02.065 --> 00:37:05.025
Composeo feeds that information back to OpenAI.

00:37:05.025 --> 00:37:10.880
So now I can give OpenAI access to all the random tools that exist if they're in the compose your directory

00:37:10.960 --> 00:37:15.440
while not giving up my password. Right? It's like the ultimate middleman

00:37:15.520 --> 00:37:22.720
in a way. Basically, it is the ultimate middleman. It's like the guy with the trench code. What do you need? And they have everything. They have a generous free tier.

00:37:22.880 --> 00:37:29.015
This is one of the tools we're going to need. This is going to be basically our main connector. So I would say,

00:37:29.415 --> 00:37:30.775
Callum, this is

00:37:31.095 --> 00:37:37.335
step number one. When you say connector, Mike, can you just make it clear for people? What is it what does that mean?

00:37:38.010 --> 00:37:38.570
So

00:37:39.130 --> 00:37:40.410
the agent

00:37:40.730 --> 00:37:41.850
can do things,

00:37:42.410 --> 00:37:44.330
but it needs access to your information.

00:37:44.730 --> 00:37:48.810
For example, let's say I want to give the agent access to my email.

00:37:49.290 --> 00:37:51.290
It needs to connect to Gmail

00:37:51.370 --> 00:37:58.265
so that it can read my emails. That's what I mean by connectors. Literally, all you do on Google is go OpenAI Codex download.

00:37:58.425 --> 00:38:02.185
The URL you're going for is openai.com/codex.

00:38:02.185 --> 00:38:06.265
I've already installed it. You can install it, and then this is what the app looks like.

00:38:06.930 --> 00:38:11.250
That being said, plug ins is basically how you connect external

00:38:11.330 --> 00:38:12.290
services,

00:38:12.370 --> 00:38:14.210
external tools to,

00:38:14.530 --> 00:38:15.250
um,

00:38:15.490 --> 00:38:21.535
Codex. Now I can see here they have one that allows me to control my Chrome browser.

00:38:21.775 --> 00:38:24.575
They have one where I can create and edit spreadsheets.

00:38:24.655 --> 00:38:26.655
I can create an edit presentation.

00:38:26.655 --> 00:38:31.135
So, again, even looking at this, I can start of already think of things like, oh, I could do my taxes, whatever.

00:38:31.830 --> 00:38:38.470
I could connect to my Notion. Right? This is pretty cool. Now let's say I wanted to connect my Notion, how I would do it? I would click the plus button,

00:38:38.870 --> 00:38:47.215
and then blah blah blah blah blah. I'm just going to tell it to reference memories in chat, meaning I want it to remember the stuff it's done. And I literally click install,

00:38:47.695 --> 00:38:50.975
and it's going to basically connect my Notion. So this is what

00:38:51.295 --> 00:38:54.735
connectors are. Yeah. So so just to make this, um,

00:38:54.975 --> 00:38:56.095
clear for people,

00:38:56.495 --> 00:38:58.655
when they start using codex,

00:38:59.270 --> 00:39:00.390
OpenAI

00:39:00.390 --> 00:39:02.390
already has plugins

00:39:02.550 --> 00:39:03.830
like connectors

00:39:03.830 --> 00:39:05.670
with a bunch of companies.

00:39:05.670 --> 00:39:08.310
Like, typically, the biggest companies, your Gmail,

00:39:08.710 --> 00:39:09.510
Notion,

00:39:09.750 --> 00:39:11.030
like things that you would use.

00:39:12.155 --> 00:39:15.355
Where Composio comes in is Composio

00:39:15.515 --> 00:39:21.195
so that is OpenAI connecting directly with this third, like this other third party company.

00:39:21.595 --> 00:39:23.915
Composio is like this middleman

00:39:23.915 --> 00:39:25.195
that you can

00:39:25.790 --> 00:39:27.470
it will have the integrations

00:39:27.710 --> 00:39:29.070
with your,

00:39:29.310 --> 00:39:30.750
yes, your Notion

00:39:31.070 --> 00:39:33.470
and maybe, I don't even know, like an Instagram

00:39:33.470 --> 00:39:34.110
or

00:39:34.510 --> 00:39:38.190
Gmail. Someone using NetSuite. Right? Like, something random. Yeah. Right? Like,

00:39:39.255 --> 00:39:43.175
they're not going to build a NetSuite integration because it's probably not popular. Maybe it is. I'm wrong.

00:39:43.655 --> 00:39:45.095
Co Compozio,

00:39:45.575 --> 00:39:54.070
their whole business model is there's thousands of tools that these companies don't care about. We're going to make it easy to connect. So if something isn't

00:39:54.070 --> 00:39:56.390
in the official plug in directory,

00:39:56.470 --> 00:39:59.510
guess what I'm going to use? I'm going to use Composeo.

00:39:59.590 --> 00:40:02.150
Yeah. So you you get access to more plug ins.

00:40:02.790 --> 00:40:05.505
Base. How how do you actually integrate

00:40:06.865 --> 00:40:08.785
how is Composeo integrated?

00:40:08.785 --> 00:40:12.065
Yeah. I'm gonna go to composeo.dev.

00:40:12.225 --> 00:40:15.185
Right? This is their website. Oh, very fancy fancy.

00:40:15.345 --> 00:40:16.465
I'm gonna create an account.

00:40:17.610 --> 00:40:18.170
And,

00:40:18.570 --> 00:40:26.410
um, when you are onboarding, they're going to give you two options. They're gonna say, is this for you? Meaning, do you wanna connect apps to your AI client,

00:40:26.570 --> 00:40:32.010
or are you building a platform? Right? Platforms for the developers looking to build on top of Composeo.

00:40:32.315 --> 00:40:39.355
What you and me want is the for you. So I'm gonna click on for you, and you're going to see it tells me, welcome back, Michael.

00:40:39.595 --> 00:40:41.435
Um, explore what's possible.

00:40:41.595 --> 00:40:45.115
You can see here it says the most used app is YouTube. I connected YouTube,

00:40:45.275 --> 00:40:49.850
um, to my coding agent, and I'm gonna show an example of how you connect apps.

00:40:50.330 --> 00:40:58.010
There's a plethora of cool things here. You can actually click on connect apps, and you can, like, look at I mean, I'm scrolling, and I'm scrolling really hard.

00:40:58.410 --> 00:41:02.250
Like, there is a lot. Right? Like, I'm still scrolling. I'm still

00:41:02.825 --> 00:41:05.225
I I I think I I even just saw

00:41:05.705 --> 00:41:10.265
Polymarket. Right? Like, if someone wants to connect their agent to Polymarket,

00:41:10.265 --> 00:41:13.705
you can. Right? But how do I make this connection?

00:41:13.865 --> 00:41:16.025
The way you do it is you click on install.

00:41:16.540 --> 00:41:23.820
Right? And it gives you actually a couple options. Do you wanna use this with OpenClaw? Do you wanna use this with Claude Cowork or Claude Code?

00:41:24.060 --> 00:41:25.820
Or do you wanna use this with Codex?

00:41:25.980 --> 00:41:29.500
And all you have to do is I'm gonna click install.

00:41:30.295 --> 00:41:36.615
Any of these work, but I'm just gonna show you what is going to work the smoothest for you. You're gonna click on MCP,

00:41:36.935 --> 00:41:38.775
and you're gonna click on API key.

00:41:39.015 --> 00:41:49.800
And then you're going to I'm going to reveal this, uh, but then I'm going to change this. If any of you pesky hackers is watching to see if you got an API key of mine, you did not. You take the same screenshot. So I take this,

00:41:50.120 --> 00:41:51.480
reveal this right here,

00:41:51.800 --> 00:41:53.400
and then I'm gonna drag this.

00:41:53.640 --> 00:41:54.440
And I'm gonna say,

00:41:55.375 --> 00:41:58.335
I want you to connect to

00:41:59.215 --> 00:42:00.175
Composeo

00:42:00.175 --> 00:42:05.775
so I can so, actually, so you can connect to external

00:42:05.855 --> 00:42:06.655
tools.

00:42:07.215 --> 00:42:10.730
I'm literally this is what you do. You hit enter,

00:42:10.890 --> 00:42:12.250
and it sets up itself.

00:42:12.490 --> 00:42:15.850
And the reason why this is possible is because the agent

00:42:16.090 --> 00:42:19.210
writes code that's going to connect to Composeo.

00:42:19.370 --> 00:42:20.570
And Composeo

00:42:20.570 --> 00:42:26.645
is built in a way where it gives you access to all these tools. Right? And to show you how

00:42:26.805 --> 00:42:27.765
this works,

00:42:28.085 --> 00:42:30.965
I'm going to go back to Composio,

00:42:31.285 --> 00:42:33.605
and let's say I wanted to connect,

00:42:34.245 --> 00:42:36.965
um, what's a good thing to connect to?

00:42:37.580 --> 00:42:40.300
Let's say I wanted to connect to cal.com.

00:42:40.300 --> 00:42:44.700
I use cal.com every now and then. I'm gonna say, I want to connect

00:42:45.660 --> 00:42:48.140
to cal.com

00:42:48.140 --> 00:42:49.820
using Composio.

00:42:50.380 --> 00:42:51.340
I'm gonna hit enter,

00:42:52.165 --> 00:42:58.245
and we're gonna see what happens. What's basically gonna happen is the agent's gonna realize, oh, I have Composeo connection.

00:42:58.565 --> 00:43:01.205
Composeo has access to cal.com,

00:43:01.605 --> 00:43:06.000
and it's basically going to give me a link to connect to cal.com,

00:43:06.080 --> 00:43:07.280
and then Composeo

00:43:07.280 --> 00:43:09.360
is going to give Codex

00:43:09.360 --> 00:43:20.465
access to it. Yeah. Let me know if that verbiage makes sense. No. It it does make sense. One one thing I wanted to point out, and it's something that I'm noticing as you're going through this, the because this is my first time actually seeing,

00:43:20.785 --> 00:43:22.065
uh, Codex

00:43:22.065 --> 00:43:23.905
and even, like, its interface.

00:43:24.305 --> 00:43:25.425
It's very

00:43:25.745 --> 00:43:26.305
similar

00:43:26.545 --> 00:43:31.665
and and you let me know, Mike. It's very similar in how it works to just using ChatGPT.

00:43:32.130 --> 00:43:34.850
I know I know most people have obviously used ChatGPT,

00:43:34.850 --> 00:43:35.890
at least in America.

00:43:36.770 --> 00:43:39.810
The functionality feels pretty simple

00:43:40.050 --> 00:43:41.090
and intuitive

00:43:41.330 --> 00:43:43.570
for someone that's used ChatGPT.

00:43:44.130 --> 00:43:50.515
Yeah. And I think what they're going to do is they're probably gonna merge the app. Like, I think Codex is the experiment on, like, developers

00:43:50.755 --> 00:44:06.160
and, like, getting them to, like, getting them to see if they like the model. And then because the Chargebee interface is pretty familiar with most people, they're probably just gonna merge the two. Right? So that's what I, again, my honest guess, I think it is. And, yeah, like, into your point, like, it shouldn't terrify you. Right?

00:44:07.120 --> 00:44:13.040
It looks the same. It feels the same. It's just a lot more powerful. Now going back to this, it says composeofoundcal.com

00:44:13.040 --> 00:44:26.415
toolkit. There's no active connection yet. It says I'm going to start an off flow. So this is basically it trying to connect to the MCP server. I'm going to allow this to happen. The cool thing is, um, these agents won't go rogue without your permission.

00:44:26.655 --> 00:44:28.095
It asked me, can I make the connection?

00:44:28.580 --> 00:44:34.660
I said yes. Now check this, Callum. It says open this Composeo link to connectcal.com.

00:44:34.900 --> 00:44:36.420
I'm going to click this link.

00:44:37.860 --> 00:44:46.175
And now what Composeo is basically doing is it's it's basically making me log in to the app. Composeo handles the credential management.

00:44:46.255 --> 00:44:50.415
My agent's going to have access to cal.com. So I'm gonna click on allow.

00:44:51.215 --> 00:44:54.895
It says successfully connected. Let's go back to

00:44:55.055 --> 00:44:55.935
Codex,

00:44:55.935 --> 00:44:59.055
and then this is the cal.com account is, uh, active.

00:44:59.490 --> 00:45:08.050
I'm doing one read only probe through Composer, so we know the co the connector works beyond just off. Now someone might say, how the heck am I supposed to understand

00:45:08.050 --> 00:45:12.850
some of these words? What you do is you copy this, you open a new chat. What does this mean?

00:45:13.835 --> 00:45:18.875
Literally, like, it and and this is how I started my career. It's like just asking,

00:45:18.955 --> 00:45:24.475
like, the instead of, like, being threatened by the difficult words, asking what they mean. And you'll be surprised,

00:45:24.875 --> 00:45:30.920
they actually are very simple. Yeah. I think it's such a great point, you know, especially what you mentioned earlier

00:45:31.080 --> 00:45:34.520
about, um, like, future proofing yourself.

00:45:34.840 --> 00:45:39.720
So many people that I speak to have this, like, uh, fear and anxiety of being, like, replaced

00:45:40.145 --> 00:45:42.385
in the in their job and in their occupation

00:45:42.625 --> 00:45:43.825
by AI.

00:45:44.225 --> 00:45:49.745
And it's like, yes, these learning how to build with these agents can be incredibly useful right now.

00:45:50.225 --> 00:45:51.025
But also

00:45:51.185 --> 00:45:52.785
moving forward in the future,

00:45:53.265 --> 00:45:56.130
that kind of, like, hustler,

00:45:56.290 --> 00:45:57.410
self starter,

00:45:57.410 --> 00:45:59.170
self taught mentality

00:45:59.490 --> 00:46:02.770
of, like, I come across something that I don't understand.

00:46:03.090 --> 00:46:04.370
I take that thing,

00:46:04.690 --> 00:46:08.690
I give it back to AI and ask it to explain it, and I learn

00:46:09.045 --> 00:46:12.565
is actually, to me, it feels like what's gonna be a key skill,

00:46:12.965 --> 00:46:18.245
uh, for people, like, moving forward in this kind of new era of AI that we're going into.

00:46:18.725 --> 00:46:19.525
100%.

00:46:19.525 --> 00:46:25.660
Because, like, you you have to think about it. Right? Like, just a couple years ago, you were sort of limited by the,

00:46:25.900 --> 00:46:26.940
uh, you know,

00:46:27.420 --> 00:46:40.795
amount of knowledge and information you have either within yourself or in in close proximity of the people you have. Now for, like, $20 a month, I have PhD level intelligence in front of me. Right? And it would be such a bummer

00:46:40.955 --> 00:46:47.675
to not take advantage of this. And even in, like, my development journey, there's a lot of technologies that I would be I would have been afraid to try.

00:46:47.915 --> 00:46:58.340
But now I'm so delusional with it. Like, I'll try new things. I'll build random things. Like, my mindset is there's nothing I can't build if I have the right information. Now I have the right information.

00:46:58.660 --> 00:47:19.995
Is it going to take time? Of course. It might be painful? Of course. I'm sure everybody watching this knows greatness takes time. Anything worth value takes time, but now I have the information. Information is no longer a blocker. Right? So that's how I would love people to take these tools. Right? Like, to like, the the this tool can help me progress and push forward. Yeah. I I I love what you said. Uh, greatness takes time.

00:47:21.080 --> 00:47:24.440
You know what? You mentioned it earlier, like, for that delusional

00:47:24.440 --> 00:47:25.080
person.

00:47:25.960 --> 00:47:30.760
For the person that's, like, high agency, self starter, just wants to make things happen, wants to build.

00:47:32.455 --> 00:47:37.575
You said that the the step one for them would actually be to build a workflow.

00:47:37.655 --> 00:47:40.215
And so I wanna talk to that person for a second.

00:47:40.855 --> 00:47:44.935
When you're thinking about workflows, I know you have, like, the sponsor agent.

00:47:45.415 --> 00:47:47.975
How can someone even decide

00:47:48.630 --> 00:47:49.830
what workflows

00:47:49.830 --> 00:47:51.350
would actually be

00:47:51.510 --> 00:47:54.150
good to have their AI agent

00:47:54.710 --> 00:47:58.950
take over? Like, talk me through your process of how you even identify

00:47:58.950 --> 00:48:00.070
the workflow

00:48:00.230 --> 00:48:02.550
that makes sense to use codex with.

00:48:03.375 --> 00:48:13.455
Back back to the diagram. So how to create how to think about workflows? This is how I think about them. Right? A, number one, anything repeated

00:48:14.015 --> 00:48:14.575
often.

00:48:15.110 --> 00:48:18.710
So if there's any task in your life that's often repeated.

00:48:18.710 --> 00:48:19.350
Right?

00:48:19.510 --> 00:48:23.670
Um, like, it's checking emails for a specific thing,

00:48:23.990 --> 00:48:27.270
whether it's, like, you know, synthesizing specific information,

00:48:27.895 --> 00:48:29.735
Anything repeated often

00:48:29.975 --> 00:48:33.175
can be and should be a workflow.

00:48:33.335 --> 00:48:34.855
Um, number two,

00:48:35.255 --> 00:48:38.215
um, and this is I would call this discovery.

00:48:40.150 --> 00:48:52.310
You know, data is a gold mine. And believe it or not, most people, because they use 15 different tools, have a lot of data about themselves. Right? Um, most of us are probably not performing

00:48:52.310 --> 00:49:01.375
at the level we want to. Right? And there's all these different things that we have to do. So the second type of workflow is what I call discovery, and that is, you know, connect,

00:49:01.855 --> 00:49:03.295
you know, every tool

00:49:03.535 --> 00:49:11.930
that is important to me. Right? Email being one of them. Notion might be another one. Um, you know, Figma, linear, whatever those tools are.

00:49:12.250 --> 00:49:14.410
And, basically, having a discussion

00:49:14.570 --> 00:49:15.770
with the agent

00:49:15.850 --> 00:49:19.370
on the stuff that you do. Right? And when I say discussion,

00:49:19.370 --> 00:49:20.090
it can be

00:49:20.605 --> 00:49:41.320
what are some actionable things that you see that I should do from all the information. Right? Like, read my emails, read my this, read my that. What are some actionable things that I can do? Um, generate a report on how much things I've done. Like, the the way I want people to think of workflow is I have all this I have this agent. Right? I have OpenAI Codex right here.

00:49:41.640 --> 00:49:43.240
I have all these

00:49:43.560 --> 00:49:53.545
pieces of information. Right? I have this right here, this right here. Uh, I have this tool right here. I have this tool right here. I have this tool right here. All these different tools that I'm using,

00:49:53.865 --> 00:49:57.305
the way I want people to think is to use this agent

00:49:57.305 --> 00:49:58.745
as a synthesizer

00:49:58.800 --> 00:50:02.320
for all the external tools they use, combine that information,

00:50:02.400 --> 00:50:08.160
and see what productive thing can be done. So a lot of it, unless you have repeated work,

00:50:08.320 --> 00:50:09.360
is discovery.

00:50:09.360 --> 00:50:11.920
But in discovery, I've noticed a lot of

00:50:13.265 --> 00:50:14.945
useful things. For example,

00:50:15.185 --> 00:50:17.105
um, I didn't know

00:50:17.345 --> 00:50:23.105
that like, when I when I'm talking about discovery, I didn't know that I was spending,

00:50:23.105 --> 00:50:28.880
um, I had a lot of emails unread. Meaning, like, I would have this problem of I would open an email.

00:50:29.280 --> 00:50:33.360
In my mind, sometimes my mind races so fast. In my mind, I've responded.

00:50:33.520 --> 00:50:44.185
I've written the response in my mind and sent it, but I'll go back and I'll do something else. And the one thing, the agent card is, like, there's a lot of opportunity in your emails that you haven't responded to.

00:50:44.585 --> 00:50:46.825
Right? Again, not like

00:50:47.065 --> 00:50:57.865
it wasn't a thought of a workflow in the moment, but I realized, oh, like, I'm really bad with my emails. Aren't that? So every week, I have a I have a a an agent. Every week, it checks

00:50:58.190 --> 00:51:02.110
all the emails that have been sent in that week. Have I missed any opportunity?

00:51:02.110 --> 00:51:04.830
Is there anything I need to check? Is there anything,

00:51:05.550 --> 00:51:10.990
uh, I need to anyone I need to respond to? And anyone who's emailed me at least twice knows

00:51:11.345 --> 00:51:20.225
at least one of those emails I responded a week after that was sent, and that's because the agent reminded me and I responded. Right? So I would think of repeated things first,

00:51:20.545 --> 00:51:38.270
and then number two is I would go discovery mode. Connect all the external tools that you use that have your inform like, that you work on, whether it be your email, calendar, whatever it is, and synthesize that information to see what can I improve, what can I automate, what things am I slacking on, what can I do better, am I do I need all these tools? Right?

00:51:39.005 --> 00:51:43.085
Think of it like a consultant. Right? Like, it's honestly a great consultant,

00:51:43.165 --> 00:51:47.165
uh, when you steer in the right direction. That's how I would think about workflows.

00:51:47.245 --> 00:51:53.920
Yeah. You know what? Two two things come to mind because, like, what you said is so good. So so the first thing is and this is what I did,

00:51:54.320 --> 00:51:55.200
is actually

00:51:55.360 --> 00:51:58.880
so you said anything that's repeated, anything that you're doing weekly.

00:51:59.120 --> 00:52:06.320
And so one thing that's helpful in the beginning is actually just starting to document what are the things that you're doing on a weekly basis.

00:52:06.885 --> 00:52:18.005
Like, either writing it down or putting it in your notes, and you'll come up with this list. It'll give you the list of just potential workflows off of that. The second thing, and I actually hadn't thought about this,

00:52:18.485 --> 00:52:20.645
and you mentioned it with the discovery,

00:52:21.125 --> 00:52:25.130
is in this digital world that we're in, we're being bombarded

00:52:25.130 --> 00:52:27.050
with so much information,

00:52:27.050 --> 00:52:28.650
whether it's like our inbox,

00:52:28.810 --> 00:52:30.010
social media,

00:52:30.490 --> 00:52:32.650
um, like text messages,

00:52:33.130 --> 00:52:41.265
that there's like and it's actually kinda scary to think about all of the things that we're not aware of, like, the thing that we forgot, but it's actually important.

00:52:41.985 --> 00:52:43.105
An opportunity

00:52:43.185 --> 00:52:47.025
that we were, like, excited about, but we let it drop off, and now it's gotten buried.

00:52:47.460 --> 00:52:50.660
And it's funny because one of the things that I'll do,

00:52:51.140 --> 00:52:55.460
and maybe this is a use case for me personally, and I'm sure for people that are listening at home,

00:52:55.780 --> 00:52:57.460
is when I'm on Instagram,

00:52:57.860 --> 00:53:00.260
I will be, like, saving posts.

00:53:00.695 --> 00:53:03.335
Like, if I see a post that could be, like, good inspiration

00:53:03.335 --> 00:53:09.335
to create content or I wanna respond to, I'll save it. I'll do the same thing on Twitter.

00:53:09.575 --> 00:53:11.175
And I just have, like, these

00:53:11.415 --> 00:53:13.735
these bookmarks full of posts.

00:53:13.815 --> 00:53:18.040
I never look at them again. Yeah. And it's like you could have an agent.

00:53:18.200 --> 00:53:21.880
I I think that's like a relatable thing. You could have an agent that's surfacing,

00:53:22.520 --> 00:53:29.160
like, learnings or things from these posts so that you can actually use them. And this is why I hate when people

00:53:29.555 --> 00:53:49.110
download or use people's specific workflows because you've now limited yourself to their box. Right? The whole point of this is for you to, like, start why and start to figure out what makes sense to you. And and in this case, like, for you, you said, um, that use case made sense. In my case, this use case made sense. Right? So it's just a matter of connect tools,

00:53:49.590 --> 00:53:58.425
figure out what to do with this data, something actionable comes out of it. Yeah. I I I think it's such a good point of, like, the whole game is making this work for you.

00:53:58.825 --> 00:53:59.545
And so,

00:54:00.265 --> 00:54:03.625
Mike, if someone's listening, I wanna go back to, uh,

00:54:04.105 --> 00:54:04.985
Codex

00:54:05.305 --> 00:54:06.345
for a second.

00:54:06.505 --> 00:54:12.970
And I'm thinking about what you mentioned in the beginning, which is, like, most people build AI agents

00:54:13.290 --> 00:54:14.490
in completely

00:54:14.490 --> 00:54:17.050
the wrong way. Like, they do it the opposite

00:54:17.450 --> 00:54:20.490
of how you should. I've even I've heard you talk about that.

00:54:21.050 --> 00:54:24.570
Like, they maximize for what looks cool over what's actually productive.

00:54:25.495 --> 00:54:26.295
Can you

00:54:26.855 --> 00:54:27.655
share?

00:54:27.655 --> 00:54:29.095
So I've connected

00:54:29.175 --> 00:54:30.055
Codex

00:54:30.055 --> 00:54:31.335
with Composio.

00:54:31.335 --> 00:54:34.135
Composio has access to all of these plug ins.

00:54:34.935 --> 00:54:35.895
What am I

00:54:36.215 --> 00:54:38.135
at high level, what am I doing

00:54:38.600 --> 00:54:42.120
so that I actually build the agent that's productive?

00:54:42.360 --> 00:54:43.320
Connecting

00:54:43.320 --> 00:54:44.280
Composio

00:54:44.280 --> 00:54:52.760
was as simple as follows. You create an account, and by the way, you can you can do this. If you're like, oh, this is too you can do this. I believe in you. You go to install,

00:54:53.475 --> 00:54:54.915
You go to codex.

00:54:55.555 --> 00:54:56.995
Click on MCP.

00:54:57.075 --> 00:54:58.515
Click API key.

00:54:58.595 --> 00:55:01.875
Reveal this. You can copy this and paste it in the chat,

00:55:02.195 --> 00:55:04.595
but then I would take a screenshot of this,

00:55:04.995 --> 00:55:07.555
and then you send it over to

00:55:08.380 --> 00:55:12.460
Codex. Right? And, basically, what I what I tell Codex

00:55:12.460 --> 00:55:14.140
is the same thing I did here.

00:55:14.780 --> 00:55:19.420
Like, um, where is it at? Right here. I basically dropped the screenshot,

00:55:19.900 --> 00:55:22.060
and I basically told it to connect to Composeo.

00:55:22.405 --> 00:55:27.045
When it connects to Composeo, this is what's going to happen. It's going to say,

00:55:27.605 --> 00:55:28.725
Composeo

00:55:29.205 --> 00:55:30.885
is configured locally,

00:55:31.205 --> 00:55:33.605
and I also asked it to connect to my YouTube.

00:55:33.765 --> 00:55:38.250
Once Composeo is connected, right, that's the first step, connect Composeo.

00:55:38.330 --> 00:55:39.290
Number two,

00:55:39.530 --> 00:55:42.730
you can go to connect apps on Composeo,

00:55:42.810 --> 00:55:45.690
and you can search if your integration exists.

00:55:45.850 --> 00:55:49.930
Right? So let's say I wanted to connect Twitter. Let's say I wanted to click connect Perplexity.

00:55:50.105 --> 00:55:54.905
Let's say I wanted to connect to Google Docs. In this case, you see I connected YouTube,

00:55:55.225 --> 00:55:56.505
and I connected

00:55:56.825 --> 00:56:08.920
Dubb. Right? Again, connecting was as easy as this. Like, I'm gonna show you another example. Like, this is the easier easiest part. Like, anyone should be able to do this. I wanna connect dub.co

00:56:08.920 --> 00:56:10.040
for my analytics.

00:56:10.200 --> 00:56:14.920
I've already connected Composeo. It's as simple as me telling it, let's connect.

00:56:15.640 --> 00:56:17.400
It gives me a link.

00:56:17.880 --> 00:56:18.760
I click the link,

00:56:19.665 --> 00:56:25.105
and then it connects it. In fact, here's the thing. And I'm glad I took a screenshot. I kept this.

00:56:25.585 --> 00:56:26.145
I

00:56:26.865 --> 00:56:27.985
clicked the link,

00:56:28.385 --> 00:56:31.505
and I went to the washroom. I had to freshen up for this interview.

00:56:31.985 --> 00:56:38.500
Um, and I took too long, and it said invalid or expired link. Now the nontechnical person is probably gonna see this in panic.

00:56:38.820 --> 00:56:50.660
What do you do when stuff like this happens? You take a screenshot and you give it to the agent. And that's exactly what I did. I said, I even apologized. I shouldn't have, but I did. I said, sorry. I took so long. It expired.

00:56:51.275 --> 00:56:53.595
And then it said and it fixed it.

00:56:53.915 --> 00:56:58.315
And I asked for it to pull some analytics,

00:56:58.315 --> 00:57:00.715
and then it did. Right? So

00:57:01.435 --> 00:57:09.670
the basically, I'm what I what what the first step is you connect composing. Second is, how do you connect further tools? You literally just have a conversation.

00:57:09.830 --> 00:57:13.270
I'll show one more example so people don't think I'm crazy. Let's

00:57:13.670 --> 00:57:14.230
connect

00:57:16.390 --> 00:57:19.030
you know what? Linear. I'll connect linear. Linear is

00:57:19.995 --> 00:57:31.755
a tracking platform for developers on to track progress, like what you've shipped, what you haven't. So I'm gonna show you how easy it is to connect Linear. I'm just gonna start a new chat, and I'm gonna say, please

00:57:31.755 --> 00:57:32.635
connect

00:57:32.795 --> 00:57:34.075
to Linear.

00:57:35.410 --> 00:57:36.690
And I'm gonna send that.

00:57:37.090 --> 00:57:38.850
And we're gonna let the agent think,

00:57:39.570 --> 00:57:41.570
and the agent is going to give me,

00:57:43.010 --> 00:57:52.335
whether it be a link or a button to connect. So this is the first thing that I want people to do is just connect the tools that you use. Don't connect the tools you want to use.

00:57:52.975 --> 00:57:53.615
No.

00:57:54.095 --> 00:57:59.535
Connect the tools that you already use. Right? So we can see here,

00:57:59.935 --> 00:58:10.260
it's saying the linear tool namespace is available. I'm going to make it read only. Oh, I've already connected it before. So it's like, linear is is connected and OAuth is working. Verified Michael Schimulus,

00:58:10.260 --> 00:58:13.940
and these are the different teams I'm running. I said it all to say connecting

00:58:14.340 --> 00:58:26.215
Calum, I hope I make sense, is literally as easy as telling the agent. You don't have to write code. You don't have to do anything. You tell the agent. You check it, compose. You'll see if it has the tool. Right?

00:58:26.375 --> 00:58:31.655
See if it has the tool. If let's say I'm a Shopify store person. There's a Shopify connection.

00:58:31.655 --> 00:58:35.170
I can literally say I can go back and say, okay.

00:58:36.290 --> 00:58:37.010
Connect

00:58:37.650 --> 00:58:39.490
connect to Shopify

00:58:39.570 --> 00:58:40.370
now.

00:58:41.090 --> 00:58:45.090
And it's the same thing. It's going to check. It's gonna fire off Composeo.

00:58:45.090 --> 00:58:48.965
Composeo is going to give me a URL. I authenticate on that URL,

00:58:49.045 --> 00:58:51.205
and now I've given eight my agent access

00:58:51.925 --> 00:58:53.045
to Shopify.

00:58:53.605 --> 00:59:01.700
Does does that make sense before I continue on? Yeah. That makes sense. So beautiful. So I'll just create a new chat. I have a couple things connected.

00:59:01.940 --> 00:59:03.380
I have my YouTube

00:59:04.100 --> 00:59:05.060
analytics.

00:59:05.060 --> 00:59:06.580
I have my YouTube analytics.

00:59:06.580 --> 00:59:08.340
I have my Dubb connected.

00:59:08.580 --> 00:59:10.900
The one thing that I do is,

00:59:11.300 --> 00:59:13.060
again, I generate these reports.

00:59:13.300 --> 00:59:17.015
Now I've held off on this tool because this tool

00:59:17.335 --> 00:59:21.015
should only be discovered by people who are serious. And if you watch to this point, you're serious.

00:59:21.255 --> 00:59:23.815
There's a tool, I don't know if you're familiar with this,

00:59:24.375 --> 00:59:26.055
column called WhisperFlow.

00:59:26.135 --> 00:59:31.230
Oh, I love WhisperFlow. So I'm using the free tier. And the reason why I love WhisperFlow,

00:59:31.230 --> 00:59:39.310
WhisperFlow allows me to voice to voice to text with the agent. So I click on the function button on my keyboard

00:59:39.390 --> 00:59:41.925
and I can talk to it. So I can say,

00:59:43.205 --> 00:59:47.925
I want to generate a report that basically tracks

00:59:48.085 --> 00:59:49.205
the performance

00:59:49.205 --> 00:59:49.845
of

00:59:50.485 --> 00:59:51.365
sponsored

00:59:51.365 --> 00:59:54.965
YouTube videos so that I can use it as marketing material

00:59:55.330 --> 00:59:56.450
for sponsors.

00:59:56.610 --> 00:59:58.530
Basically, I do a lot of content

00:59:58.770 --> 01:00:05.490
and you have a connection so you can see my content and some of the videos are sponsored and you can see the sponsored links

01:00:05.650 --> 01:00:06.850
in dubb.co

01:00:06.850 --> 01:00:14.285
and there's a connection for that. I wanna generate a a report that basically or some sort of marketing material that I can send to sponsors

01:00:14.285 --> 01:00:16.765
so they can send me their hard earned money.

01:00:17.805 --> 01:00:19.965
Now typing that would have taken me forever.

01:00:20.580 --> 01:00:22.420
But what's cool about WhisperFlow

01:00:22.420 --> 01:00:35.295
is there's a lot of information here that sometimes it's hard to distill via typing because it takes too long. I will voice the text the heck out of this. Another thing that you mentioned, Calum, that I'll add and I'll just hit enter with this is

01:00:35.615 --> 01:00:43.455
you said, um, to voice to text your day, your thoughts, and I would 100% do that. Right? You can use Notion, some people use Obsidian,

01:00:43.695 --> 01:00:51.850
really documenting your work, right, via voice to text and then feeding that into an agent, you'll be surprised how lazy we are naturally,

01:00:52.010 --> 01:00:58.490
um, will help you a lot. Right? All these things are very simple, but when you put them together,

01:00:58.730 --> 01:01:02.275
you start to become very, very, very, very productive.

01:01:02.275 --> 01:01:03.475
I think talking,

01:01:03.795 --> 01:01:08.835
a, is a skill that everyone should have. So, I mean, using WhisperFlow almost, like, forces you.

01:01:09.235 --> 01:01:14.115
Um, but b, I find, at least for me, I get a lot of information out of myself

01:01:14.660 --> 01:01:16.420
when using WhisperFlow.

01:01:16.580 --> 01:01:20.900
Um, that being said, I typed in this giant prompt and this agent

01:01:20.980 --> 01:01:23.860
is added. Right? It checked Composeo.

01:01:23.860 --> 01:01:31.255
It checked the tools. It notices I have 50 recent uploads. It notices that I have 33 links, and now it's trying to match,

01:01:31.495 --> 01:01:34.535
um, the videos. Now this will probably fail.

01:01:34.855 --> 01:01:41.095
Now the reason why this will fail is I gave it too broad of a task. If it succeeds, I'll be surprised.

01:01:41.335 --> 01:01:49.360
If it fails, I'll understand why. And I I'm low key hoping it fails, but if it doesn't fail, then it's just proof that the agent is smart.

01:01:50.480 --> 01:01:58.080
But, essentially, what I did was not a good prompt. Yeah. My my my question to you, because you said I think that's really interesting. You said that you gave it too broad of a task.

01:01:58.975 --> 01:02:01.935
In your mind, kinda take people behind the scenes.

01:02:02.895 --> 01:02:04.735
What is the signal to you

01:02:05.055 --> 01:02:08.255
that you've given the agent too broad of a task?

01:02:08.495 --> 01:02:20.870
Okay. So give me sixty seconds of your time because you're gonna want to hear this. So I've done over 200 episodes of this show. I've been doing it for four years. In the last six months, I started to notice a shift in the conversations.

01:02:20.870 --> 01:02:23.750
Things that the guests would talk about was different.

01:02:23.830 --> 01:02:32.275
The common message was that we had entered this new period of time where the rules to be successful and have a great business and a great life had fundamentally

01:02:32.275 --> 01:02:44.770
shifted and changed. And so I started to realize that we had entered a new era. And so a couple of weeks ago, we launched a secret newsletter called New Era that pulls together stories,

01:02:44.770 --> 01:02:46.210
supporting documents,

01:02:46.290 --> 01:02:57.455
step by step guides, even unseen footage from the show, and we send it to you in your inbox every single week. And so if you wanna be involved, just go to the link in the description,

01:02:57.695 --> 01:03:02.175
plug in your email, and you will get an issue of our new era newsletter

01:03:02.175 --> 01:03:04.335
every single Monday in your inbox.

01:03:04.670 --> 01:03:15.550
Thank me later. Let's get back to the show. I'm glad you asked, and I have an example for that because I was hoping you're gonna ask it. So I basically, in this prompt said I wanna generate a report that basically tracks this. This is

01:03:16.875 --> 01:03:18.395
Pretty pretty general.

01:03:18.795 --> 01:03:21.755
But if I go to the last,

01:03:23.595 --> 01:03:26.875
where is it? Is this one right here, I want to show you

01:03:27.355 --> 01:03:28.475
one interesting thing.

01:03:29.260 --> 01:03:31.420
I'm gonna show you the steps I took it.

01:03:31.900 --> 01:03:37.500
So I first connected the tools. Right? So you can see here I gave him my API key

01:03:37.660 --> 01:03:40.220
and then it says it's configured.

01:03:40.780 --> 01:03:45.595
I told it asked it tell me what my last video did in views. This is a very simple prompt.

01:03:45.755 --> 01:03:49.275
I want my last video and I want the views. Can't get this wrong.

01:03:49.915 --> 01:03:55.355
It says it's currently sitting at 12,000 views, 800 likes, 44 comments.

01:03:55.675 --> 01:03:56.475
Simple.

01:03:56.555 --> 01:04:05.700
Right? Now the reason why I told it to do this is this action is in the context window. Meaning, uh, it's in the it's in the it's in the mind of the agent.

01:04:06.740 --> 01:04:10.980
The next thing I tell it is, let's connect to Dubb, which is my analytics platform.

01:04:11.780 --> 01:04:13.220
Gives me the URL.

01:04:14.345 --> 01:04:18.665
Remember, I I went to the washroom to fix my hair. Took too long. I said sorry.

01:04:18.905 --> 01:04:24.505
But then it connected it. And then look what it did. It says, I verified a sponsor link from your latest video.

01:04:24.745 --> 01:04:30.990
Because I asked it to check my video first and it knew that I wanted the analytics for some sort of analytics,

01:04:31.150 --> 01:04:34.110
it checked the link that matches.

01:04:34.270 --> 01:04:37.630
And it says here, oh, here are the current results. 534

01:04:37.630 --> 01:04:38.430
clicks.

01:04:38.590 --> 01:04:39.630
This is the destination.

01:04:39.885 --> 01:04:43.725
This is the last click. I'm not tracking leads or sales or sales amount.

01:04:43.965 --> 01:04:49.965
And then it did it itself. It's like combined with the YouTube number from earlier, the current sponsor

01:04:50.125 --> 01:04:51.485
CTR is

01:04:51.565 --> 01:04:53.085
4.43%.

01:04:53.405 --> 01:04:56.600
Right? Now this prompt is going to feel lazy,

01:04:57.000 --> 01:05:05.880
and I'm going to tell you why it works. It's and I said, yes. Now do me a favor. Generate a clean looking page showcasing this number. This is for a sponsor.

01:05:06.425 --> 01:05:10.025
I'm allowed to be lazy with this prompt because

01:05:10.025 --> 01:05:13.145
I've given it context with the previous conversation.

01:05:13.545 --> 01:05:15.065
Right? It's like

01:05:16.345 --> 01:05:22.170
I don't have a perfect example, but it's like leading someone on. Right? First, you're like, you know, I really like your shoes.

01:05:22.650 --> 01:05:33.930
And then you're like, I really like your t shirt. And you're like, I really like your smile. At some point, you're gonna understand, yeah. This person really likes me. Right? That's essentially what I've done. Right? I first gave it access to my YouTube

01:05:34.330 --> 01:05:38.645
video views, and then I I gave it access to my analytics,

01:05:38.885 --> 01:05:40.245
and now I'm telling it

01:05:41.045 --> 01:05:42.485
generate a report.

01:05:42.725 --> 01:05:44.485
And what it does is

01:05:45.285 --> 01:05:49.445
and I oh, and here's the thing. I I I tell it to generate a clean looking page.

01:05:50.130 --> 01:05:54.210
Here's why I do it. When it generates this page, it's going to write HTML.

01:05:54.450 --> 01:05:55.330
HTML

01:05:55.330 --> 01:06:04.690
is, uh, is basically a file type. It's what a lot of websites render. This file type is great for agents. Agents excel with HTML and markdown files.

01:06:05.915 --> 01:06:07.675
I told it to generate this.

01:06:07.995 --> 01:06:09.675
If I click on this right here,

01:06:10.155 --> 01:06:16.235
you can see that it generated this report. You know, not too shabby. If I don't like the style, here's what's cool about Codex.

01:06:16.315 --> 01:06:21.780
They have an annotation tool. I can click on this annotation tool. I can select something,

01:06:22.180 --> 01:06:24.820
and then I can tell it I don't like this font. I don't like this

01:06:25.780 --> 01:06:31.145
k? So I can I can go pretty, like, specific with this, but I don't care?

01:06:31.705 --> 01:06:40.345
It generated this report and I liked it. Now here is where you don't download people skills, but you create your own. We had a successful run.

01:06:40.585 --> 01:06:43.065
So you know what I did, Callum? I said, see,

01:06:43.940 --> 01:06:49.060
this right here is a great report. And you could tell I kinda talk it to talk to like, I talk to homies.

01:06:49.220 --> 01:06:56.100
I say, can you turn it into a skill where I just tell you the video and you generate these pages?

01:06:56.820 --> 01:07:06.715
So I've given it what I want the end result to be, and I've told it what input it can expect from me. What input it can expect from me is a YouTube URL.

01:07:06.795 --> 01:07:09.915
What I expect from it is the is this right here.

01:07:10.315 --> 01:07:12.955
And this took a second. This took five minutes.

01:07:13.640 --> 01:07:16.600
And here's one thing I want people to realize. Sometimes,

01:07:17.080 --> 01:07:31.465
the agents will take time. And there have like, for example, in this instance, it was waiting for me to give it access and I didn't. So, right, I have to click allow and it's probably gonna keep on working. But going back to this example, and this is why I generate these examples ahead of time,

01:07:32.745 --> 01:07:35.145
because this video would probably been way too long.

01:07:35.945 --> 01:07:37.945
You could see it took its time.

01:07:38.265 --> 01:07:47.760
I could maybe work on another task. I could maybe do another thing. Again, this isn't my full time job. I have other work to do. But while this is happening, I could go do something else. After five minutes,

01:07:48.080 --> 01:07:52.000
it created you could see here, created the reusable skill.

01:07:52.800 --> 01:07:57.360
This is very, very important. The reason why this is important is the next time

01:07:57.925 --> 01:07:59.525
I want this done,

01:07:59.605 --> 01:08:01.525
which I have another example for you,

01:08:02.005 --> 01:08:03.285
I can tag it.

01:08:03.525 --> 01:08:04.085
So

01:08:04.325 --> 01:08:06.485
I can click at on

01:08:07.125 --> 01:08:08.485
the text box

01:08:08.565 --> 01:08:10.405
and you can see a bunch of these

01:08:11.070 --> 01:08:16.670
tags. Like, it it it it it it wants me to tag one of these things. All I have to do is click generate

01:08:16.750 --> 01:08:20.190
and you can see this skill. This is the very skill

01:08:20.510 --> 01:08:21.870
I created.

01:08:22.030 --> 01:08:29.595
So I created this skill and what I'm gonna do is I'm gonna show you actually an action. Yeah. Okay. This video was sponsored. I'm gonna copy this link,

01:08:30.555 --> 01:08:32.235
and I'm going to paste it.

01:08:32.715 --> 01:08:34.475
So let me paste it here.

01:08:35.915 --> 01:08:39.275
Remember, I told the skill earlier. Let me go back.

01:08:41.910 --> 01:09:00.305
I told her right here when I generate the skill, can you turn this into a skill? What is it turning into a skill? It's turning the workflow. Basically, the step by step I took. You see a lot of people, Callum, will tell the agent, I want you to do this, I want you to do that, I want you to do that. Turn this into a skill. Wrong.

01:09:01.505 --> 01:09:03.425
Have a successful run first

01:09:03.665 --> 01:09:05.665
because sometimes it will fail.

01:09:06.145 --> 01:09:18.530
Have a successful run first. Once you've had the successful run, because that conversation is in the AI's context, is in the agent's context, you're just going to turn it. We had a successful run. Turn this into a skill.

01:09:19.330 --> 01:09:20.850
These are the best skills.

01:09:21.330 --> 01:09:40.390
The best. The ones that you generate after a successful run. I'll stop here because I think you're about to say something. Yeah. Yeah. I am. I love how passionate you are about this. It's it's it's good, and and it's so useful. And it's interesting the the point in time that we're speaking at because, like, I'm learning a lot of this stuff right now. For people that are listening and might be unaware,

01:09:40.630 --> 01:09:41.910
can you explain

01:09:42.630 --> 01:09:44.470
what is a skill

01:09:44.710 --> 01:09:47.190
in, like, layman layman's terms?

01:09:47.590 --> 01:09:48.550
And then what

01:09:48.870 --> 01:09:50.630
if I'm looking to be more productive,

01:09:51.065 --> 01:09:55.145
make more money, save time, what is the value of using Skills?

01:09:55.465 --> 01:09:56.425
Skills,

01:09:57.465 --> 01:09:58.585
they're they're

01:09:58.585 --> 01:10:00.905
noted as skill dot m d files.

01:10:01.305 --> 01:10:06.010
What's interesting about Skill files is they will they will contain instructions

01:10:06.010 --> 01:10:08.090
on how to do a specific task.

01:10:08.490 --> 01:10:10.490
But here's what's interesting about it.

01:10:11.050 --> 01:10:13.930
The specific task, like the full task,

01:10:14.250 --> 01:10:17.850
the instructions of the task are not fully given to the model.

01:10:18.170 --> 01:10:20.675
Right? I'll even go a step granular.

01:10:21.235 --> 01:10:23.794
Every skill file contains the following three,

01:10:24.115 --> 01:10:24.915
the name,

01:10:25.235 --> 01:10:26.195
the description,

01:10:27.315 --> 01:10:28.435
and the steps.

01:10:28.835 --> 01:10:29.475
Right?

01:10:30.675 --> 01:10:32.275
The name is the name of the skill.

01:10:32.870 --> 01:10:43.350
The description is it describes what the skill allows you to do. The steps are whether it's code or a step by step instructions, that's what the agent follows to do set task.

01:10:45.075 --> 01:10:50.435
What's given to the agent when a skill is created is just the name and description.

01:10:50.995 --> 01:10:58.515
Why is this important? The reason why this is important is we don't wanna fill we don't wanna fill the mind of the agent with too much information.

01:10:58.770 --> 01:11:03.890
Remember, if you give it too much information, it's gonna be overwhelmed the same way you're overwhelmed

01:11:03.970 --> 01:11:05.730
and it's not going to perform well.

01:11:06.130 --> 01:11:06.930
Skills

01:11:06.930 --> 01:11:14.785
are a trick. Like, we don't wanna dump in every single instruction that we have and this is why skills are so important. It's minimal information

01:11:14.865 --> 01:11:16.785
that's called progressively,

01:11:16.785 --> 01:11:22.145
and when it's called, it gives all the steps. And all this is to make the agent more performant.

01:11:23.025 --> 01:11:26.065
Hopefully, that that makes sense. Yeah. So it sounds like it's

01:11:26.620 --> 01:11:28.060
minimal information

01:11:28.380 --> 01:11:39.820
that is still guiding the agent. Imagine having, um, this book and I have this book has all the information here, but all I know and that's stored in my head is is the chapters.

01:11:40.245 --> 01:11:53.925
And if someone tells me, oh, such and such chapter, I don't need to know it offhand. I'll just open a book, go to the chapter and read the specific thing. That's what the model does with skills. It stores the name and the description. It doesn't store the steps.

01:11:54.405 --> 01:12:16.985
But when it knows it needs the steps, it then reads it in. This makes it minimal. And you can think of this almost like a a computer, you know, the less the more memory you have, the more space you have in your computer, the faster it is. Right? The more stuffed it is with files and the closer it is to its memory, you start to see it slugs and it's slow and it lags. That's basically what skills do for models.

01:12:16.985 --> 01:12:24.745
It just gives it enough information it needs to know when it needs it. And when it needs it, it calls the entire thing. Yeah. No. That's really clear.

01:12:25.600 --> 01:12:27.120
I love that. So

01:12:27.760 --> 01:12:29.520
that's what was generated here.

01:12:29.840 --> 01:12:46.215
Now let's go back to the prompt I typed in here, uh, this one right here. I said generate a sponsor report. Right? I tagged the skill. Right? And how you tag skills is you type an at and you just type the name of the skill. In this case, this skill was named generate sponsor report.

01:12:46.695 --> 01:12:52.695
And I tagged and I gave it the YouTube video, and it says I'll generate a second report. I'm fetching the YouTube metadata.

01:12:52.775 --> 01:12:57.350
It's asking me does it have access to call these tools? I'm going to say yes.

01:12:57.590 --> 01:13:01.590
And what's going to happen is it's going to pull in the information.

01:13:01.910 --> 01:13:05.430
Notice how it's asking for another request. I'm going to say yes.

01:13:06.255 --> 01:13:14.095
And, again, if any of these lines don't make sense to you when you're using it, you copy this, you create a new chat, you paste, and you get answers.

01:13:15.775 --> 01:13:22.970
And now it has the metrics. You can see here, I have the live metrics, 15,000 YouTube views, 279

01:13:22.970 --> 01:13:24.170
tracked clicks.

01:13:24.330 --> 01:13:29.930
I'm creating a JSON payload, rendering the page as a standalone HTML page. So it has the information.

01:13:30.170 --> 01:13:32.330
It's now going to create this page.

01:13:33.185 --> 01:13:38.305
And in just a little while, the HTML page is rendered. I'm checking the output.

01:13:38.865 --> 01:13:46.225
Agents do take time. The smarter they are, the longer they will take. It says it's generated here. If I click, hopefully, this works.

01:13:47.790 --> 01:13:50.749
And there you have it. I have a repeatable

01:13:51.150 --> 01:13:51.709
process.

01:13:52.270 --> 01:13:55.070
Now let's say I wanted this done daily.

01:13:55.790 --> 01:13:56.910
What do I do?

01:13:57.230 --> 01:13:59.630
I go in my chat. I open up WhisperFlow.

01:14:00.795 --> 01:14:05.515
Thank you for generating this report. Can you make sure that this happens

01:14:05.915 --> 01:14:14.635
every single week? I want a report generated every single week for the latest video. I do upload week weekly so it makes sense. Uh, make that happen.

01:14:16.440 --> 01:14:17.800
And I hit enter.

01:14:18.760 --> 01:14:21.080
And now the model has a skill.

01:14:21.560 --> 01:14:23.560
It has access to my YouTube channel.

01:14:23.800 --> 01:14:29.480
What I've essentially created Calum right now is a weekly workflow where reports going to get generated.

01:14:29.720 --> 01:14:34.975
And notice how it's it's so simple. It's literally conversation.

01:14:36.735 --> 01:14:43.855
It did it right here. So it says set up a weekly latest sponsor report. It will run every Thursday, 9AM American Toronto time.

01:14:44.015 --> 01:14:47.790
And if I click on this right here, I can see information.

01:14:48.110 --> 01:14:48.830
The the

01:14:49.230 --> 01:14:51.230
OpenAI calls them automations

01:14:51.390 --> 01:14:52.510
to sound fancy.

01:14:52.830 --> 01:15:09.865
Um, and look what it it generated. It tell me it tell me here, generate a sponsor facing performance report for the latest public YouTube upload on the connected RossMike channel. Use the generate sponsor report workflow. Remember, we generated it earlier. Fetch the latest video metadata and statistics.

01:15:10.025 --> 01:15:16.985
Identify the sponsored dub short link from the video description. Fetch the dub link info and click counters.

01:15:17.250 --> 01:15:19.890
Calculate track link CTR,

01:15:19.970 --> 01:15:22.050
then render a standalone HTML

01:15:22.050 --> 01:15:31.825
report plus a JSON data payload, name the files with the sponsor name and the video ID so each weekly run keeps its own output. Keep conversion

01:15:31.825 --> 01:15:33.185
caveats factual.

01:15:33.185 --> 01:15:39.025
Do not expose API keys, OAuth tokens, and account IDs or raw tool logs.

01:15:39.185 --> 01:15:43.025
This is going to run the next one it says here, 05/21/2026

01:15:43.025 --> 01:15:58.430
at 9AM, the status is active. This is going to happen every single week. Now how I built this, I didn't see someone on YouTube build this and copy them. I built this on based on what's useful to me. So connect your tools. You can use OpenAI's native plug in marketplace.

01:15:58.590 --> 01:16:02.725
If the tool doesn't exist there, which most likely won't, you connect Composeo.

01:16:03.045 --> 01:16:06.165
Composeo is as easy as going to the dashboard,

01:16:06.325 --> 01:16:08.165
taking a screenshot of the instructions,

01:16:08.405 --> 01:16:10.565
giving it to Codex, and saying connect.

01:16:10.885 --> 01:16:22.140
It will do it. And then after that, ask yourself what tools do I use? Ask the chat. I want to connect to this tool. Every time you ask to connect to a specific tool, it will give you a link, you will authenticate,

01:16:22.140 --> 01:16:34.805
it will bring you back and now the agent has access to it. Then you can enter discovery mode. I want you to do this, I want you to do that, I want you to do that. And when you realize a couple of the steps that you did, you want it to be done consistently,

01:16:34.965 --> 01:16:39.845
what are you going to do? You're going to turn this. You're going to tell it itself, turn this to a skill.

01:16:40.485 --> 01:16:45.285
Once it's a skill, it's become a workflow. And if you want it to happen on a consistent basis,

01:16:46.040 --> 01:17:02.985
nothing fancy here. I literally said thank you for generating this report. Can you make sure that this happens every single week? And guess what? It set up an automation and it does this every single week now. So on May 21, next week, 9AM, I'm going to get a message from Kodak saying here is the latest report.

01:17:03.545 --> 01:17:12.025
Yeah. You know you know what, Mike? It's so good, and and thank you for sharing that. And the reason I say it's so good is because of what you mentioned in the beginning, which is that it's simple.

01:17:12.680 --> 01:17:18.040
It's not 50 different agents in these, complex workflows and demos.

01:17:18.120 --> 01:17:23.880
It's very simple. And so when you and I spoke, I actually wrote down three steps,

01:17:24.280 --> 01:17:26.040
which is basically step one,

01:17:26.705 --> 01:17:32.145
you you're using a codex, you connect it to composeo.

01:17:32.305 --> 01:17:36.464
Step two is pick a real task that you do every day or every week,

01:17:37.585 --> 01:17:46.200
and you coach the model step by step. So you go through that workflow step by step with the model first so that you get,

01:17:46.600 --> 01:17:48.280
like, a great output.

01:17:48.600 --> 01:17:51.640
Then once you've got a great output from the agent,

01:17:52.505 --> 01:17:56.024
you then tell the agent to turn that into a workable skill,

01:17:56.105 --> 01:17:59.145
and then you put it on some sort of repeated cadence.

01:17:59.785 --> 01:18:03.705
You know what, Mike? So I'm gonna ask a a selfish question here.

01:18:04.265 --> 01:18:04.985
Because

01:18:05.385 --> 01:18:06.745
back to my experience

01:18:07.440 --> 01:18:11.680
of using Claude Cowork and, like, reflecting on it, because it was a frustrating

01:18:11.760 --> 01:18:13.840
and, like, a demoralizing experience.

01:18:14.160 --> 01:18:18.560
Because I spent hours setting it up, and then the output that I got

01:18:18.800 --> 01:18:26.975
wasn't useful. And the funny thing is, it told me at first, it's like, yeah, it won't be great at first, but it'll get better over time and all this stuff.

01:18:27.455 --> 01:18:29.295
And what happened in my experience,

01:18:29.615 --> 01:18:32.015
because I didn't get a great output to begin,

01:18:33.180 --> 01:18:38.060
I didn't really use it moving forward because I knew I wouldn't get a great output.

01:18:38.540 --> 01:18:41.340
And so when we talk about the steps,

01:18:41.740 --> 01:18:43.580
I think that step two

01:18:44.140 --> 01:18:44.780
of

01:18:45.355 --> 01:18:46.875
people understanding

01:18:47.035 --> 01:18:49.995
kind of how to coach the AI agent

01:18:50.315 --> 01:18:51.275
initially

01:18:51.355 --> 01:18:53.435
to give them a great output,

01:18:53.835 --> 01:18:56.795
that is the key step. Because once you get the great thing,

01:18:57.490 --> 01:19:01.090
then it's so easy as you've just demonstrated to turn it into a skill.

01:19:01.650 --> 01:19:03.730
But I'm curious for you,

01:19:04.690 --> 01:19:06.690
can you talk about your experiences

01:19:06.770 --> 01:19:07.410
of

01:19:07.970 --> 01:19:10.290
getting a mediocre output

01:19:10.515 --> 01:19:14.515
or like a suboptimal and average unusable output

01:19:14.675 --> 01:19:16.115
from this process?

01:19:16.995 --> 01:19:23.715
And then what are you doing to actually make it good so that further down the line you can then turn that process into a skill?

01:19:24.920 --> 01:19:26.280
Yeah. So

01:19:27.240 --> 01:19:29.880
I, uh, grew up in an African household,

01:19:30.120 --> 01:19:37.080
uh, which means if my grades were an a plus, I got b. Right? And some might call this craziness,

01:19:37.245 --> 01:19:43.885
but now I look back at it and this was just a great feedback loop. I do good, I don't get beat. I do bad, I do get beat. I

01:19:43.965 --> 01:19:46.205
bring that example up because

01:19:46.365 --> 01:19:48.285
you are going to have runs

01:19:48.445 --> 01:20:04.340
where again, I'm gonna use this diagram so it's super, super clear. You're going to have runs where you are just going to be frustrated because what the agent gave you, what you you gave it a good prompt, what feels like a good prompt, and what it gave you was pretty mediocre.

01:20:05.835 --> 01:20:09.115
What you are going to do is you are going to

01:20:09.675 --> 01:20:12.235
continue to work on it recursively

01:20:12.555 --> 01:20:20.490
as if it was an annoying family member that you're stuck with but are frustrated with. What do I mean by this? Most of the times,

01:20:20.730 --> 01:20:31.290
the agent will probably mess up. Right? Um, the models are getting better so luckily it worked out on our end. But let's say it messed up. And in this case, there was a perfect example

01:20:31.450 --> 01:20:36.955
of actually not even it messing up, but I think me taking too long. If I go back here,

01:20:37.355 --> 01:20:40.475
I took too long and it expired.

01:20:40.555 --> 01:20:46.075
And I know for a fact most people would see this and would just panic and would probably give up.

01:20:46.720 --> 01:20:52.160
What I want you to do and this is depending on the task might be time consuming

01:20:52.320 --> 01:20:59.315
is you recursively work with it. Meaning, you tell the output that it's giving you is wrong and it needs to itself.

01:20:59.715 --> 01:21:13.395
So let's say tries to connect to something and it feels connecting. You're going to tell it you failed at connecting at it, please fix this. You'd be surprised by the third or fourth time it figures it out because each time we'll probably try a different strategy doing so.

01:21:14.140 --> 01:21:23.740
When it comes to creative work, this is probably the most painful because in creative work, um, the sauce is not trained into the model. Right? Your specialty,

01:21:23.740 --> 01:21:25.660
your flow, your cadence,

01:21:25.980 --> 01:21:37.465
and I know there are a lot of people that are like, oh, I'm fine tuning a model so it sounds like me. Um, humans are very unique in where a lot of the things we do are very subtle and are very hard to replicate,

01:21:37.625 --> 01:21:39.145
uh, by a machine.

01:21:39.465 --> 01:21:40.025
So

01:21:40.500 --> 01:21:42.420
when it comes to stuff like that,

01:21:42.740 --> 01:21:48.980
the feedback loop that you're going to give it is the same loop that like my parents did. You do good,

01:21:49.460 --> 01:22:23.705
we create a skill. You do bad, we continue to work with you until you create a skill. It's literally just back and forth and and this is why I'm saying like the models have gone really good. The models are so intelligent now where if you tell it this was the wrong way and then you show it the right way or you go back and forth and it does it the right way and you create a skill, guess what happens? In the skill, it will write, I tried these different ways and this was wrong. The person didn't like it or it didn't work. Never use this method again. Right? So step two to your point and you put it perfectly

01:22:23.705 --> 01:22:29.545
is the most important and it's probably the time you spend the most. Now, if it feels frustrating,

01:22:29.625 --> 01:22:44.480
I want you to think of this as a human that's being trained on the job and you would have a little bit of compassion with a human. Hope at least you would be nice to someone who just started working for you. In the same way, I'm not saying be nice to the model but understand there might be a little training that's required.

01:22:45.040 --> 01:23:05.625
Once you have that back and forth and the back and forth is simple, it's just prompting. It's just communicating with it. Something breaks, you tell it it broke. Now let's say compose you. Let me give you an example of how I would fix an issue. Let's say Composio went down and now Composio is just no longer a company. I would go on the agent and it'd be like, I want you to search the web and find a greater Composio alternative and

01:23:06.770 --> 01:23:21.810
instruct me on how to set it up. I would go then set it up and if it breaks, again, I would continue to prompt back and forth and that's one thing I want to encourage people. If something isn't working, prompt the AI. If that isn't working, prompt AI.

01:23:21.970 --> 01:23:24.895
If Codex isn't working, prompt Claude.

01:23:25.295 --> 01:23:30.975
Prompting is where you fix a lot of the issues you might get in a workflow that's not working correctly.

01:23:31.135 --> 01:23:37.775
For some of you, it might be one shot, which praise God that works. But for some, it might take a couple of tries.

01:23:38.550 --> 01:23:39.590
Hang in there.

01:23:39.990 --> 01:23:45.830
Continue to work with it. Once it has a successful run, you're going to have the best skill of life.

01:23:46.230 --> 01:23:46.950
Yeah.

01:23:47.110 --> 01:23:54.150
No. Mike, the the analogy of, like, the the African or, like, the immigrant parent is, very vivid

01:23:54.895 --> 01:23:55.615
to me.

01:23:57.215 --> 01:24:08.575
But you know what? It's it's it's a it's a great one. And so I wanna kinda go behind the scenes of your process for a second. So you get, like, a bad output from your agent, and and you correct me if I'm wrong in this.

01:24:09.450 --> 01:24:10.410
You're

01:24:10.650 --> 01:24:13.450
hitting whatever that key is that activates.

01:24:13.450 --> 01:24:23.290
So you're like studying the output first of all. Like, you're looking at what it's given you. You're then hitting that key that gets you to whisper flow and you're like dictating

01:24:23.290 --> 01:24:23.770
out

01:24:24.245 --> 01:24:27.524
what was wrong or what you want corrected

01:24:27.605 --> 01:24:28.965
in the output,

01:24:29.285 --> 01:24:31.445
you're doing this process

01:24:32.005 --> 01:24:32.885
repeatedly

01:24:32.885 --> 01:24:35.765
until you actually get to a good output.

01:24:36.100 --> 01:24:40.820
And then the instant that you get to a good output, you're like, okay. Save that workflow

01:24:41.060 --> 01:24:42.500
as a skill.

01:24:42.900 --> 01:24:44.340
Is that correct? That's

01:24:44.340 --> 01:24:53.705
that's exactly it. That's the that's the step by step summarized perfectly right there. Right? Because it has all the perfect inform it has all the right information in context.

01:24:53.865 --> 01:24:56.825
I just want it to be turned to skill as soon as possible.

01:24:57.145 --> 01:25:08.890
Yeah. You know what? One one final thing, Mike, because you've been so great. When you set it up on this reoccurring basis and you walk people through how they can do it so it actually you know, every week, it delivers this.

01:25:09.690 --> 01:25:11.530
Where is that output

01:25:11.610 --> 01:25:12.890
going to,

01:25:13.610 --> 01:25:28.455
and do you control that? So for example, are these sponsor reports I don't know. Before you even send them out, is it going is it like an email that it's sending you? Is it, uh, uploading it onto, like, your computer, like, a local drive?

01:25:28.935 --> 01:25:29.655
Um,

01:25:29.895 --> 01:25:33.575
where is it going typically for you, and what have you found

01:25:33.895 --> 01:25:35.975
works best when you're using an agent?

01:25:36.630 --> 01:26:26.605
So when you use something like Codex, Codex can do all of the things you just mentioned. Right? So I can have it just pop up in the chat. Right? Like like, just message me in the chat when something's done. Because if Codex is important to my business, then probably have this thing open or I'm using it regularly. So it could do via chat. I can also tell it to email me. Right? Again, all it would take is and this is what's so cool. All it would take is for me to connect my email and tell it, oh, send it to me. Right? Alert me. Um, there are even tools that can allow you to connect it to a telegram or to all this stuff and how do you like, how would I go about setting this up? Check this out. Prompt it. I want you to send this information to my telegram or to this or to that and it will do it and that's the cool part about these apps. So I say that to say, if you want the information

01:26:26.685 --> 01:26:27.485
delivered

01:26:27.565 --> 01:26:39.770
to you via email, tell it. You want it in a specific folder in your file system, tell it. You want it uploaded on Google Drive? Tell it. It'll tell you to connect and you could do it. You want it sent to you via telegram?

01:26:39.770 --> 01:26:42.250
Tell it. How you want things delivered

01:26:42.410 --> 01:26:53.315
is completely and utterly in your fingertips. The power is in your hands. It's just as simple as prompting it. Yeah. And it's so good. What have you found is what's best for you?

01:26:54.435 --> 01:27:00.515
I'm a I'm an iPhone guy and I really like Apple. I I I really like iMessage.

01:27:01.395 --> 01:27:02.595
So I have,

01:27:03.155 --> 01:27:04.500
like, my own

01:27:04.980 --> 01:27:08.340
I I can show it here, but, like, this is, like, my own agent.

01:27:08.580 --> 01:27:27.555
But I think if you're starting out, I would really focus on using the app and getting used to the app because the app has so much functionality. Like, another thing, and this is just a side quest that people can do is you can type at and I believe I can click like build. Yes. I can tag build iOS apps,

01:27:27.555 --> 01:27:30.035
which is like a plugin they have, and I can say,

01:27:30.835 --> 01:27:36.140
build me a to do list app that looks like, uh, the Apple native reminder.

01:27:36.220 --> 01:27:39.740
And I also want you to add light mode, dark mode toggle.

01:27:40.460 --> 01:27:42.860
Right? And then I can hit enter,

01:27:43.180 --> 01:27:53.235
and I can let this work. I can work on another task. Right? I think there's value in getting used to the app, right, especially if you're new. Focus first on the app.

01:27:54.115 --> 01:28:01.629
Get good at it. Use it. Tinker around. Click all the random things. Touch all the settings. Like, just, you know, use it profusely.

01:28:01.710 --> 01:28:12.110
And then once you've got to a good stage, then you can worry about, okay. I can connect it to my iPhone to this to that. Because I find that a lot of people will do all the unproductive stuff that's cool

01:28:12.445 --> 01:28:17.405
and then be like, oh, I just wasted so much time. Like, all of this was kind of, like, useless.

01:28:17.565 --> 01:28:19.645
Right? But if we scale for productivity,

01:28:19.965 --> 01:28:21.885
make the productive thing first,

01:28:22.205 --> 01:28:29.245
then it just becomes something you're stuck with. Like, I can't stop using agents because they make me so productive. Right? So that's

01:28:28.210 --> 01:28:39.730
how I would think about, like, where to use and how to use the things. I would stick to the app first. Yeah. I'm curious for you, reflecting on kind of what you've experienced and getting to the point that you are now,

01:28:41.105 --> 01:28:47.985
what would you say to the person that's listening? If they had to take one thing from this conversation

01:28:48.065 --> 01:28:51.745
and even really just from hearing your story and experience,

01:28:52.145 --> 01:28:53.345
what would you hope

01:28:53.585 --> 01:28:55.585
is the one thing that they would take from you?

01:28:56.890 --> 01:28:57.930
Yeah. Like,

01:28:58.170 --> 01:29:03.690
I I genuinely think all the information you have is available. Right?

01:29:04.650 --> 01:29:06.490
It's there. It's on YouTube.

01:29:06.650 --> 01:29:09.450
It's on Twitter. It's on it's it's it's all there.

01:29:09.690 --> 01:29:15.625
Right? What is important is consuming information and then acting upon it. Right?

01:29:16.585 --> 01:29:17.945
A lot of people,

01:29:18.345 --> 01:29:18.985
um,

01:29:19.625 --> 01:29:44.365
will consume and not do. And I've done this before. I've been like a tutorial junkie before where I'll watch a bunch of tutorials on how to do x y and z. And it and there is a dopamine rush in learning something new, uh, but that can also be a mistake because you want to actually do. Right? So I would consume information and do. I would consume information and do. And I found that especially in my own personal experience after a certain amount of time,

01:29:44.605 --> 01:29:46.365
at some point, you become

01:29:46.445 --> 01:30:00.830
better than what's average. Right? Like, I remember in August 2022, I decided to take, um, Harvard had this has this free course called the CS 50, which is like an introduction to computer science program. And it's like an eight week program.

01:30:01.150 --> 01:30:04.510
And it was probably the most, like, humbling,

01:30:05.230 --> 01:30:07.310
borderline humiliating

01:30:07.755 --> 01:30:10.955
thing. And by the way, look, the app is being built itself while we talk.

01:30:11.355 --> 01:30:19.915
That's so cool. Yeah. It it it was very humbling and and, like, I almost felt embarrassed because I I I just didn't feel I could keep up. Like, it was so hard. It was so difficult,

01:30:20.315 --> 01:30:29.350
but I stuck stuck to to it it and and that that led led to me building my first few apps. I sucked at it. I stuck to it. That led me to starting a startup.

01:30:29.750 --> 01:30:39.365
That startup did really, really well. I stuck to it. But then I had a feud with a business partner. I ended up leaving the company with no equity, no cash,

01:30:39.525 --> 01:30:50.860
no money even though that business was doing like couple million dollars a year. I stuck to it. Right? And then I got my first corporate job. It was annoying, but I stuck to it. Right? So there's just like this consistent

01:30:51.100 --> 01:30:53.660
nature of like sticking to

01:30:53.900 --> 01:31:28.540
this idea of learning and improving, learning and doing, learning and improving, learning and doing. And it never happens right away, but you'll realize at some point whether it's a year, six months, two years, at some point you'll realize, oh, you're a lot more smarter than you used to be and you're a lot more valuable than you are. And like that's the big thing. Right? Like when a lot of people like, oh, these tools are scary. This is even if it feels like a technical tool, so what? Try it. You know what mean? So what? You you don't know how to cope. So what? Give it a try. Use AI. Ask. Fail. Try it and if it sucks, it sucks. Right? But at least try.

01:31:30.125 --> 01:31:32.205
Another thing I'll say is and, like,

01:31:33.325 --> 01:31:37.645
like, a lot of one thing that grounds me is, like, I my faith grounds me.

01:31:37.885 --> 01:31:43.885
You know, I I have motivation. I have a family I need to take care of. Right? So aligning all these things together

01:31:45.140 --> 01:31:48.980
makes it so that, like, yeah, there's some things that I don't understand.

01:31:49.700 --> 01:31:59.460
So what? There are lot of people who don't understand. Let's go learn it. Let's invest time. Whether it's buying a book, whether it's watching some videos, whether it's trying it out, whatever it is, whether it's paying for somebody's time,

01:32:00.625 --> 01:32:05.905
the information's out there. It's just a matter of who's going to do the work, and unfortunately,

01:32:05.905 --> 01:32:12.305
most people won't do the work. Mike, man, thank you so much for coming on and being so open and not gatekeeping.

01:32:14.120 --> 01:32:16.120
I really enjoyed this conversation.

01:32:16.440 --> 01:32:19.880
I I and I appreciate you having me and, you know, for sharing.

01:32:20.280 --> 01:32:31.535
Just to show everybody, I know we're closing up, but, like, yeah, the the the app's done and, like, that was one prompt. So, like, you have access to this. There should be no reason why you don't try. At least try.

01:32:31.855 --> 01:32:40.175
But, yeah, Calum, I really appreciate the opportunity and the time, and, yeah, thank you for for allowing me to be here. Yeah. It's pretty crazy that, like, as as we're talking,

01:32:40.990 --> 01:32:42.749
the the AI,

01:32:42.750 --> 01:32:49.390
like, the the agent is just building the app and it's, like, popping off on the screen. Uh, that's cool. Yep. That's cool.

01:32:49.950 --> 01:32:51.390
Thank you, mate. That was great.

01:32:51.710 --> 01:33:03.935
Appreciate you, Callum. Thank you. So So if you enjoyed this conversation and you wanna hear even more stories like this, then just click here. And also my team is gonna put some more videos that you can watch here. Thank you.