WEBVTT

00:00:02.880 --> 00:00:05.520
I led tech for a start up that sold for $23,000,000

00:00:05.520 --> 00:00:09.120
and I've been using AI coding tools daily since.

00:00:09.120 --> 00:00:21.805
I wanna talk about how Augment's new task feature is very overpowered and probably shakes things up for a lot of companies. As a bonus, they gave me credits to give the audience which is really cool.

00:00:22.045 --> 00:00:27.325
So we're about to cover three things. One, the details of this new task feature.

00:00:27.645 --> 00:00:48.615
Two, how I think it's actually gonna kill off a lot of the AI tools in the space, and three, we'll go through my workflow. Stick around to the end of the video to learn about the credits. Like many of you, I've been overwhelmed by the sheer volume of tools and changes and updates that have been coming out. But I've found solace in sticking to a maximum of three tools, cursor,

00:00:48.775 --> 00:01:05.920
Claude code, and then augment's been a mainstay because of their context engine. If you wanna learn more about that, I have other videos on my channel. But this new task thing saves me so much time. It is just great. So why does it still rip? Let's just show you. Let's say you have a set of tasks

00:01:06.000 --> 00:01:12.640
and you want to get really specific about them and you've maybe watched some videos on somebody using

00:01:12.720 --> 00:01:20.955
a CLI tool or you've come up with your own prompts on how it maps to your Jira, your Notion, or wherever you're working out of. It might look like this.

00:01:21.195 --> 00:01:23.035
And this was like awesome

00:01:23.115 --> 00:01:30.470
two weeks ago. And you had to go in and make these very specific tasks, and you wanted to have as much

00:01:30.790 --> 00:01:39.190
specificity in there as possible, and you would prompt prompt prompt your way. Maybe you'd go over to Gemini and repo prompt because we need more tokens.

00:01:39.350 --> 00:01:56.975
You'd be optimizing for this. And this is like a decent example. And this is that I built. I was that guy who built their own pull framework around this which is just kludge. And I said that when I built it. A lot of these tools will get replaced and that's what's happening right now with what we're talking about. So do you want this?

00:01:57.215 --> 00:01:59.380
And now, you just have this.

00:01:59.540 --> 00:02:03.780
Oh, what's this little guy over here? It's bananas. What it does,

00:02:04.180 --> 00:02:34.400
and I'll show you in my code base in a second, but a lot's happening in here. You basically have your task list which you might have hid on the left side and one of these little notepads. I've seen people do that. I don't do that. You might have made a bunch of prompts and try to fling yourself around and do it that way. But you don't have to do that anymore. You just have this right here. Now, there are ways to turbocharge it, of course. But this primitive, it's clear as day. Throw it in the IDE.

00:02:34.640 --> 00:02:46.575
That's brilliant. You can swap between file change mode, which if you're familiar augment is when you see that running log of all the things that are going on. Or your task manager. If I wanna switch

00:02:46.815 --> 00:02:56.690
into task mode, I can do it in one of two ways. I can either go and manually type in the tasks. That's loser mode. That's not mode. But typically, you'll have

00:02:57.090 --> 00:03:05.170
the tasks or the things that you think you need to do before you got to this step. So if that's you, then you can literally just drop it into

00:03:05.410 --> 00:03:06.850
the chat.

00:03:06.850 --> 00:03:10.050
And then you flip on auto and hit go, and it will

00:03:10.695 --> 00:03:22.215
automatically separate those out for you. Now, benefit of doing it this way is you can take advantage of the enhanced prompt. You always want to check the outputs of LLMs. They are probabilistic

00:03:22.215 --> 00:03:23.095
geniuses.

00:03:23.095 --> 00:03:30.720
So they might not nail it as hard as you want it to nail it. That's on you. Go read what the enhanced prompt thing does. Always read your outputs.

00:03:30.880 --> 00:03:45.975
It's just it's gonna pay you in spades. But leverage it. So you drop in whatever that task list you need to do here, that's step one. Step two is hit the magical icon, let it do its sizzle, let it work its sauce, and then it will actually

00:03:46.215 --> 00:04:03.460
give you even better context because context is king. And then you can send it off to the auto agent, it will get to work and actually create the list for you. So, good example of this was I just did this and I was like, I have to record a video because it's too good. It's actual sauce. What happened was, I needed

00:04:03.700 --> 00:04:04.340
to

00:04:04.660 --> 00:04:07.860
beef up the schemas. You need Zod validation

00:04:07.940 --> 00:04:13.955
because your types don't exist at runtime, right? Like you see the little errors and the squigglies if you're familiar with TypeScript.

00:04:14.035 --> 00:04:40.125
But you don't get that same thing when you're building an API and it's out in the world. So what I needed to do was make these Zod schemas for all the different routers, routers, all the functionality that exists within VAI, which is our private network of builders, a bunch of nerds basically. Wanted to bring San Francisco and that vibe onto the Internet because I live in a tough desert. So that's the community building platform. That's what the router are for. Needed the Zod schemas.

00:04:40.285 --> 00:05:06.210
Great. I don't wanna write all those and I know how to. I can go. I can do that, but I don't want to. And you don't want to either. You maybe write one of them, but no. You don't wanna write all of them. So in my case, I always pair it with sequential thinking. I think that's one of the best mcps if you're gonna use an mcp. That's a good one. But I'll go ahead and I'll ask it specifically to make tasks for me. I obviously asked it to refine the prompt. And then you can see as I scroll down,

00:05:06.530 --> 00:05:07.090
you have

00:05:07.795 --> 00:05:09.315
the updated prompt.

00:05:09.475 --> 00:05:32.320
And then it goes and reads the directory. It's building the context around what it's about to deliver. Reading the lines, showing you what's going on. Seeing that some of them are done, some of them are not. And then boom, add task 21. You can see if it updated any existing ones. So, it's doing all that right in front of you and then we can get off to the races where I can literally just click run all tasks. It will go to completion.

00:05:32.480 --> 00:05:41.255
Couple other pro tips when you have a lot of tasks. How do you deal with it? Well, they got filtering here, which is pretty nice. So I can click on the little filter.

00:05:41.495 --> 00:05:49.815
Guess what? It does a filter. Could you did were you able to ascertain? Were you able to figure that part out? But, yeah, you can filter it. And then also,

00:05:49.975 --> 00:06:23.710
if you want to kick it in into a new chat or if you just have so many that you wanna kick it out into a markdown file, not sure, maybe you wanted to sync that with the GitHub MCP or integration, then you could. But, yeah, those two very, very helpful. Now, side note. You see this, Claude code. I wanna talk about that too. Because I'm not just using one. Right? Like I mentioned at the beginning of the video, I try to stick to three max. Cursors, the home base, it's still got the tabby thing. That's great. And then I have Claude code which is like, alright, let's jump in and let's rip around and have some fun in this.

00:06:23.950 --> 00:06:25.150
But I

00:06:25.390 --> 00:06:34.805
I gotta say, it can just get you lost really fast because you get like excited about the fact that you have all these things. And I find that it burns through the context window too quickly,

00:06:35.125 --> 00:06:41.365
and it doesn't have a grip on the code base like augment does. So while I just rode the dragon,

00:06:41.365 --> 00:06:48.730
which was Claude code for the last few days, like just literally trying to push it as hard as I could, it caused a lot of bugs that

00:06:48.810 --> 00:06:53.770
skill issue. Probably a skill issue, but no. I mean, it literally sim linked an envar,

00:06:53.770 --> 00:07:03.610
an environment variable file to somewhere else and I didn't even notice it. And then it didn't notice it either. It's I just find that this can be almost like less

00:07:03.435 --> 00:07:17.355
pointed and it knows more. It just seems smarter. That's all. What about cursor? Well, cursor, I just think it's in a different bucket. I think maybe we're at a peak. I don't wanna say it. I don't wanna say it. I just find that every day that goes by,

00:07:17.675 --> 00:07:33.855
there's so much stuff that they have to be doing and I know they're orchestrating a lot of different models underneath the hood, but it's not nailing my use case for one area of coding, is like actually doing the tasks. And I don't wanna go and make these task lists and even little things like, if you compare the way that this

00:07:33.855 --> 00:07:36.335
can output stuff versus

00:07:36.975 --> 00:07:40.015
this, like it didn't I don't know. It just doesn't nail

00:07:40.335 --> 00:07:43.215
markdown a lot of the time. It doesn't have mermaid

00:07:43.215 --> 00:07:46.070
built in. MCP's are less reliable.

00:07:46.230 --> 00:08:13.495
Any cursor watchers, this is great for you. This is an opportunity. This is you can transcribe this chucking it into the road map and throw remote agents on it. Just command e it. Right? And so now, let's get into the next segment. We're at eleven minutes already. Wow. Here I am. I thought I was like the next star. I thought I was gonna get through this a lot quicker, but I'm now I'm in Figma. Oh, here we go. Okay. If you don't follow me on x, I post once every three years. But these are pretty good memes. Right? Because this is me, literally.

00:08:13.735 --> 00:08:14.980
Like that is me.

00:08:15.300 --> 00:08:22.740
I was like I need a local l n and I have these seven prompt chains and they're tied to the AIS DLC which is the software development life cycle and it's

00:08:22.820 --> 00:08:31.615
there's elements to it that still definitely makes sense. I'm rethinking a lot of stuff around this where the Opus Research

00:08:31.855 --> 00:08:36.095
agent is bananas and that should be used as part of specs.

00:08:36.335 --> 00:08:59.885
And then some of the stuff from Indie Dev Dan, I think is really good. I just need to simplify it. But yeah, this is what I'm wondering. What's this what's going on here? Do we know? Do we know what's go how look at those eyes. I don't know. I didn't say it. You said it. Did you say it? So now let's talk about the AI tool extinction event. Yeah. Of course, unfunded, but I think that there's something to be said with looking at

00:09:00.125 --> 00:09:03.005
these tools and being realistic.

00:09:03.005 --> 00:09:09.405
Let's be realistic together for a second. Who's gonna win over the long term? I think it's the ones that have deep partnerships

00:09:09.405 --> 00:09:10.125
with

00:09:10.285 --> 00:09:31.805
foundational models. Right? So OpenAI, first in the space, crushing it. I think Claude is way better at coding across the board. I think they're better at writing. So I've just I don't even find myself using ChatGePity, to be honest, or any of their models at the moment besides transcription and embeddings. But I should use Gnomic. But I think they're fine. When I look at a tool that does not have

00:09:32.285 --> 00:09:41.390
a deep partnership like Augment and Claude does, for example, that makes me not wanna use it. Because then your pricing gets jacked up, you are

00:09:41.710 --> 00:09:57.085
not winning. Now, I'm not gonna name them. We don't name names here, but you can think of them. And so, you can go and augment all around. If you're price conscious, that's the obvious answer because you get I wanna say 600 chats for $50.

00:09:57.405 --> 00:10:04.365
So run the math on that and the chat is literally the thread. Now, quad code, if you wanna ride the dragon, it's $200

00:10:04.445 --> 00:10:10.060
and it just seems like a bull in a China shop. I like it, but probably not as productive.

00:10:10.380 --> 00:10:17.100
You feel like you're being more productive, but yeah. And so, comparing it to ClawCode Max, the Clawd Dogging problem.

00:10:17.260 --> 00:10:19.260
Yeah. I already talked about that nightmare.

00:10:19.500 --> 00:10:23.115
And so, I was gonna end this by just like walking you through

00:10:23.355 --> 00:10:24.875
another example,

00:10:25.035 --> 00:10:44.920
but I kinda wanna just talk about how I'm thinking I'm gonna work for two different work streams. What is a work stream Parker? I don't know. It's a cool word that I just said, but it also means how you will go do a set of tasks for a given goal. So let's say if I'm going one to one point n.

00:10:45.080 --> 00:10:47.960
Yeah. We're throwing n around. We're so mathematical.

00:10:48.040 --> 00:10:49.240
Then in this case,

00:10:49.640 --> 00:10:50.855
I would probably

00:10:51.255 --> 00:10:53.335
do the Auggie

00:10:53.655 --> 00:10:54.455
context.

00:10:54.775 --> 00:11:02.215
So I'm asking a lot of questions about how I need to accomplish this task. Now for any work stream, it's the old,

00:11:02.690 --> 00:11:13.730
it depends. Because if you don't know the code base like the back of your hand, you don't know the libraries, you don't know how things are done, I'm not really sure with state management because I'm new here. Then, yeah, you're gonna have to spend a lot more time upfront.

00:11:13.890 --> 00:11:22.825
But if you do know it like the back of your hand and you're committing code to a daily, then you probably don't need a lot of deep research

00:11:22.905 --> 00:11:39.610
and Auggie context questioning over and over. So probably get away with a thread. Especially if it's something that you've already built in a different before. But I'd be going in a simple state like this. I'd be like, okay. A, as I go through with the context, and then b, coming up with some sort of plan,

00:11:39.850 --> 00:11:41.530
and that's actually like

00:11:41.690 --> 00:11:46.175
maybe something like a PRD. And then I can throw that PRD PRD into

00:11:46.175 --> 00:11:47.615
AI slash

00:11:47.855 --> 00:11:48.575
specs,

00:11:48.895 --> 00:11:50.895
and then I get into tasks.

00:11:51.055 --> 00:11:58.815
And that's literally it. Now, this does not mention the fact that you should have a dot augment guidelines.

00:11:58.815 --> 00:12:23.345
How about that handwriting? I had to take extra lessons on handwriting. I'm lefty. Okay. But can you play tennis lefty and righty? Probably not. Man, that's tough. I'm tough. Okay. So this is something that I'd be doing where it's okay. This is how we do data fetching. This is how we do state management, this is how we do our types, this is how our REST API layer works, this is how our tRPC layer works. All those things kinda get outlined. And there are additional artifacts.

00:12:23.425 --> 00:12:30.840
You can put them in the guidelines if you'd like. But I really think that the beauty of this is really just, hey, this is how I like to make PRDs.

00:12:30.920 --> 00:12:33.240
This is how I like to make tasks.

00:12:33.480 --> 00:12:46.325
And this is where we store them. All those things. Whereas if I were going zero to one, and again it depends. But if I'm going zero to one, then I'm probably throwing in another step before that.

00:12:46.565 --> 00:12:47.765
So it's more Oh,

00:12:48.085 --> 00:12:48.885
opus

00:12:48.965 --> 00:12:50.565
for research.

00:12:50.885 --> 00:13:06.750
Now, if you don't have that, sorry, you should get it. It's tight. Claude, send me a shirt, would you? But you could also use ChatGePety, of course, or Gemini. And this just is you doing research, figuring out what it is your input for this. So let's do input

00:13:06.750 --> 00:13:07.550
equals

00:13:07.630 --> 00:13:20.535
your code base. So you can literally just dump it in there and then ask much questions around, hey, here are the things that we want to accomplish around the problem. You can ask to get two steel man at because you're probably coming up with a solution that's overkill, over engineered.

00:13:20.695 --> 00:13:47.195
You don't need to. So you can really do the mental jousting there. And then output would be the researched spec. So it looks something like that. Right? And then you go on to b c d. So that's where I'm thinking currently as of June 18. But yeah, very excited about all the Auggie stuff. What about the credits, dude? Oh, wow. Yeah. My bad. Okay. So, if you want the credits, make a comment below as to why you want to be using Augment.

00:13:47.515 --> 00:14:00.340
Give me a specific use case of how you plan on using it And that's it. Literally, just comment it below and I'll pick those out by the end of the next week. See you. If you want to learn more about what we're doing with the engineers inside of VAI,

00:14:00.420 --> 00:14:08.905
with people from Microsoft and Google, love to be able to say that. We have a new platform coming out. Price is gonna double because it's not a little school emoji,

00:14:08.905 --> 00:14:17.705
I'm your guru thing. It's like an actual like network with a bunch of tools in it that I'm excited about. So you can collaborate on projects in there. You can share different resources.

00:14:18.300 --> 00:14:23.900
We have a prompt library. There's gonna be public access for some of it too for discoverability

00:14:23.900 --> 00:14:38.525
purposes and also just because I think it's a great way to build a brand. So that's it for today. If you learn one thing, then you got like the video so that more people see this. And then if you learned two things or you didn't learn anything at all, you should subscribe. Alright. I'll see you in the next one.