WEBVTT

00:00:00.000 --> 00:00:15.235
Many people think that ClaudeCode is better than Cursor. So I brought on an AI coding expert, Ross Mike, to break down what ClaudeCode is and why it's changing the world of AI coding. We talked about how ClaudeCode is different than tools Cursor and Winsor.

00:00:15.395 --> 00:00:18.595
We talked about how to get started building with ClaudeCode.

00:00:18.755 --> 00:00:22.835
We looked at an app that Mickey built fully with ClaudeCode.

00:00:22.835 --> 00:00:29.555
And we also talked about why Anthropic is building ClaudeCode the way they are. Let's dive in. Ross Mike,

00:00:29.940 --> 00:00:43.460
in your last YouTube video, you said that Claude code is the best programming agent, and I've actually heard other people say the same. Do you really think that it is better than tools like Cursor? I think so. I believe so.

00:00:44.180 --> 00:00:45.460
Not only have I

00:00:45.925 --> 00:00:55.205
used it, like, very much in the last two weeks, but I also understand why I think that way. And I wanna kind of explain if it's okay with you, share my screen

00:00:55.445 --> 00:01:13.020
and sort of, like, just get into Excalidron diagram why I think so. More than anything, one thing that we all understand is that Claude is the best model for programming. Right? I think we all understand that all the big providers have sort of picked their lane. Like, OpenAI seems to be going for the more personable

00:01:13.365 --> 00:01:14.485
b to c

00:01:14.805 --> 00:01:15.605
feely,

00:01:15.605 --> 00:01:17.445
you know, human companion

00:01:17.445 --> 00:01:28.240
type vibe, and then Anthropic has picked its lane where it wants to go into the programming lane. And I actually was at a Claude code conference where they announced Claude code. And the big difference really is

00:01:28.560 --> 00:01:29.760
if you're using

00:01:29.920 --> 00:01:30.880
Cursor,

00:01:30.880 --> 00:01:31.760
Windsurf,

00:01:31.760 --> 00:01:43.155
zed, like, whatever it is. Right? I know everyone has their camps. Like, we have Cursor here. We have Windsurf here. And then let's say you're, you know, an OG and you you're using Versus Code. Right?

00:01:44.675 --> 00:01:47.315
If you're using any of these,

00:01:47.635 --> 00:01:48.675
like, tools,

00:01:49.155 --> 00:01:56.620
you notice that some people are like, oh, I think Cursor's better. Some are like, oh, I think Windsor's better, and some are like, oh, they're the same. I just use Versus Code.

00:01:57.180 --> 00:01:58.060
Underlyingly,

00:01:58.060 --> 00:02:16.515
they all use the same model. Now the question is what really makes them different? Why could one user think Windsurf is better than one user thinking Cursor's better? Now there is some sort of anecdotal point to it where, like, maybe someone just had a good run, and they think it's better, and someone had a bad one with the other tool. But, really, the the difference is

00:02:16.995 --> 00:02:19.235
they're all using the same model,

00:02:19.635 --> 00:02:21.075
but they're all building

00:02:21.075 --> 00:02:22.035
their own tools.

00:02:22.460 --> 00:02:33.980
Right? So the model in and of itself is is useless. Right? It's just an image it's just a resource store. It has all these all this knowledge. Right? So you can ask it questions, but if you tell Cloud four,

00:02:34.220 --> 00:02:42.785
read this file in my code base and tell me what it does, it has no way of doing so. So what makes Cursor, Windsurf, and Versus Code

00:02:43.105 --> 00:02:52.785
unique is they have tools that they've built. Right? So for example, like, you use Cursor and you you look at the tools it calls, like, you'll see something like

00:02:53.660 --> 00:02:54.700
file reader.

00:02:54.780 --> 00:03:00.620
Right? And then, like, you or you'll see something like I don't know if you've ever, like, seen, like,

00:03:01.180 --> 00:03:07.845
Cursor or Windsurf when they're about to push an update. Sometimes, like, you see gibberish. If you look at the gibberish, it'll be, like, writing file.

00:03:07.845 --> 00:03:09.765
Right? So you'll see

00:03:10.085 --> 00:03:10.725
that

00:03:11.525 --> 00:03:17.285
what makes Cursor and Windsurf different is that although they're using the same underlying model,

00:03:17.605 --> 00:03:26.910
they're developing their own tools. Right? So Cursor has its own file reader tool, and the way this works is maybe I can, like, illustrate this better is,

00:03:27.070 --> 00:03:28.830
let's say this user,

00:03:29.150 --> 00:03:30.750
right, goes to

00:03:30.990 --> 00:03:36.105
Cursor. Right? This user is using Cursor, and this user asks cursor,

00:03:36.185 --> 00:03:40.105
cursor, can you change the color of this page to green?

00:03:40.265 --> 00:03:53.380
What happens is cursor fires that off to Claude four o, and Claude four o is gonna say, oh, the user wants me to change the color of this page. I need to know where the page is. So what Claude is going to do is Claude is going to fire off

00:03:53.540 --> 00:03:54.580
the tool

00:03:54.660 --> 00:03:57.540
that Cursor has provided for Claude,

00:03:57.620 --> 00:04:25.780
and that tool is going to read the file, give the contents back to Claude. Now Claude is then going to be like, okay. This is where the file is. This is the code that I need to change. Now I need to change it. So Claude then fires off the other tool, and this tool allows it to write to the file. Right? These tools are developed by Cursor, Windsurf, Versus Code. And how they've implemented these tools, how these tools work together, the efficiency of the tools, how they handle context memory,

00:04:25.860 --> 00:04:28.065
this is sort of the secret sauce

00:04:28.385 --> 00:04:36.385
that Cursor and Windsurf are fighting for, which is why if you notice, Windsurf developed its own model and Cursor raised a bajillion dollars,

00:04:36.465 --> 00:04:53.290
and they're probably gonna develop their own models. Because right now, what they're fighting over is this. This is this their differentiator right now because everyone is using Claude four. Right? But what happens is when they have their own models now, they can have their own models that are trained a certain way, that becomes an edge for them. That's a business podcast.

00:04:53.705 --> 00:04:54.345
But,

00:04:54.585 --> 00:04:55.545
underlyingly,

00:04:55.545 --> 00:05:02.025
this is how Kerser and Winsthorf work. They have tools, memory, different ways of handling context.

00:05:02.105 --> 00:05:06.105
Like, they'll even maybe fine tune specific models for certain things. And,

00:05:06.825 --> 00:05:16.150
you know, working at a an AI startup, I know this. It's not just one giant Claude agent. They probably have multiple agents, maybe even using different

00:05:16.790 --> 00:05:22.070
models. Right? Maybe they have, like, a specific model and they use Gemini,

00:05:22.150 --> 00:05:32.005
like, 2.5 flash because it has a large context window that ingests the entire code base and gives, like, summarized context to Claude four because Claude is expensive.

00:05:32.405 --> 00:05:33.925
I say that to say this

00:05:34.405 --> 00:05:38.910
is the edge that Kursor and Windsurf have. Now where does Claude

00:05:38.910 --> 00:05:39.630
code

00:05:40.190 --> 00:05:42.670
come into play? Claude code,

00:05:42.990 --> 00:05:46.910
although not a similar interface, meaning they didn't fork Versus code,

00:05:48.110 --> 00:05:49.950
Claude code is essentially

00:05:50.590 --> 00:05:50.990
built

00:05:51.415 --> 00:05:53.495
on Claude's model. Right?

00:05:53.895 --> 00:05:57.495
But they, this time, have built their own tool calling.

00:05:57.815 --> 00:06:03.175
And this is sort of the understanding I have. Like, I spoke with some of the team at Claude at the conference,

00:06:03.415 --> 00:06:05.735
and first and foremost, they're going all in on this.

00:06:06.370 --> 00:06:08.290
Second of all, they told me 95%

00:06:08.290 --> 00:06:18.850
of their engineers use Claude code. Very few are using Cursor and Windsurf. The other 5% are just rebels who want to use, like, VIM or, like, maybe they'll use, like

00:06:19.330 --> 00:06:21.090
like, you know, Cursor or Windsurf.

00:06:21.545 --> 00:06:28.345
But majority of their developers are using Cloud Code. It was an originally an internal tool. Now they've made public.

00:06:28.745 --> 00:06:33.065
And the thought process I have is the one developing the model

00:06:33.225 --> 00:06:47.370
will probably know the best way to develop the tools. Right? And this is why from my personal usage again, I don't have, like, data, like, intrinsic data for you to say, oh, this is the percentage. But for my personal use, especially with larger code bases,

00:06:47.530 --> 00:06:50.010
Clawr code seems to understand

00:06:50.330 --> 00:06:52.475
code better and execute

00:06:52.475 --> 00:06:54.955
code better, and it does feel more agentic.

00:06:55.035 --> 00:07:11.180
Right? You can allow it to continue on. Like, I think it's called auto mode or something. You enable it. And not only will it continue till the task is task is complete, but there are certain instances where it will try to write tests, and it'll test the code. Right? It does feel a lot more agentic

00:07:11.580 --> 00:07:17.260
than Cursor and Windsurf and Versus Code. And Cursor launched with one point o background

00:07:17.340 --> 00:07:40.740
jobs or background agents. I think that's what they call it. And that's what I would say is a direct competitor to Cloud Code. So the reason why I think Cloud Code is is better is because the model provider is finally building the tools that make tools like Cursor and Windsurf better, and who better built it than the one actually developing the model. So that's kind of my spiel and my thesis as to why I think ClaudeCode is better.

00:07:41.220 --> 00:08:13.780
When you use ClaudeCode, do you use it within Cursor? I've heard some people use ClaudeCode in Cursor. And so, like, is that just, the ultimate stack? Because, like, I tried to use ClaudeCode, and I actually asked Cursor how to set it up. And I actually got it set up, but I didn't have the normal IDE. So it was hard for me to go to the right file, like because I was, like, uploading images, and I was trying to give it context of the file that I wanted to reference, but, like, there was nothing to look at. So can you talk about why people use Claude code within Cursor? And then, like, why you think this like, what do you think of this interface

00:08:13.815 --> 00:08:15.975
of Claude code versus cursor?

00:08:16.215 --> 00:08:18.855
It's definitely very geared to developers.

00:08:19.175 --> 00:08:23.735
And my actually favorite setup is don't know if you can see my cursor

00:08:23.895 --> 00:08:24.615
is

00:08:25.175 --> 00:08:27.415
the way I kinda have it set up is I will have

00:08:28.220 --> 00:08:28.860
Claude

00:08:29.100 --> 00:08:41.980
like, I'll open Claude code, and and I I can show, like, how, like, you set it up. But I just wanna give you, like, a quick rundown on how I have it set up is I'll pin Claude code to the top left,

00:08:42.615 --> 00:08:46.775
and then I'll still have my terminal here, and then I'll have my cursor

00:08:46.855 --> 00:08:51.655
chat. And what I've really been enjoying is I'll use cursor

00:08:51.815 --> 00:09:18.435
like, I'll use Claude four o for asking. So, basically, like, I'll build my understanding of what it's doing, how this code base works, or I'll build, like, my questions. Like, this is, like, the assistant I ask questions, and Cloud Code is sort of my executor. So it it is true in a sense where I think, especially for a Vibe coder, it's best to mingle the two. I don't think you should use Cloud Code by itself with, like, a Versus Code or just raw dog it on terminal. I think it becomes very confusing.

00:09:19.235 --> 00:09:24.355
But if you couple it with, let's say, your Windsurf guy, Windsurf and Cloud Code, or you're,

00:09:25.155 --> 00:09:43.005
you know, a cursor person, cursor and Cloud Code, I think you get the best of both worlds. And here's here's the best part. I I know there's a lot of compare comparison videos, but you can use both, right, which to me seems like a great compromise. Right? I can use cloud code. And if it feels like it's not my type of thing, I can use,

00:09:43.325 --> 00:10:06.230
like, Cursor Chat. But I think what makes Cloud Code special is the model provider is building the tools. And I don't know if I've just been on a hot streak, but I've had instances where, like, the regular, whether it be cursor or windsurf just would not get it done. Quad, even though it took longer, got the task done. And the thesis that I'm going with is the same thing I shared is I think the

00:10:06.550 --> 00:10:18.895
model provider building the tools for the model, they just end up knowing what works best. Right? So I highly, highly recommend if you're gonna give Cloud Code a shot, definitely couple it with your favorite ID, whether it be Cursor or Windsurf.

00:10:19.215 --> 00:10:23.535
Because you have to remember, I mean, these companies are hedging. Right? Like, Cursor and Windsurf,

00:10:24.060 --> 00:10:38.140
at least up until a few months ago, are building their models to be are building their product to be model agnostic, to use any of the models. Now with Windsurf, it's interesting because Windsurf is rumored to be bought by OpenAI for $3,000,000,000

00:10:38.140 --> 00:10:40.215
and now is Anthropic

00:10:40.215 --> 00:10:42.535
cutting them off of tokens,

00:10:42.535 --> 00:10:44.535
uh, intentionally or unintentionally.

00:10:44.535 --> 00:10:49.015
I don't wanna get into the rumors. I don't think it matters that much. But, like, if you think about it,

00:10:49.495 --> 00:10:50.055
it

00:10:51.415 --> 00:10:56.940
makes sense that the the company like, if you really think the edge is in the tools

00:10:57.020 --> 00:10:59.580
that it uses, like, whatever

00:10:59.980 --> 00:11:10.925
like, if quads ends up being the best coding model in the world, you would think that Anthropic would be the best at building those tools around it because they can build the model specifically for those tools.

00:11:11.565 --> 00:11:19.885
Exactly. Exactly. Yeah. They they have the And it They literally have the cheat code. And then it just makes sense for them to go all in at that point, because if they can own programming,

00:11:19.885 --> 00:11:21.405
like, that seems

00:11:21.490 --> 00:11:28.690
you know? And that that makes sense to me that Claude just owns the programming side, and OpenAI owns the kind of the the the become

00:11:29.010 --> 00:11:37.675
your best friend personality side or whatever hardware they're building. So that's really, really interesting. Kursor and Windsurf, I think they are starting to see that

00:11:38.235 --> 00:11:56.700
Claude is sort of inching in the direction where at first, there was a big separation. Right? Claude was like, we'll just provide the model. You two battle it out with UX. You two battle it out with actually getting users. Now it seems like Claudel's stepping in that front and Kerstro and Windsurf. You also wanna I think the moat for them becomes the model. Right?

00:11:57.180 --> 00:12:21.890
V zero launched, like, their own specific model that's trained on Next. Js and stuff. And I think, like, for tools like this, if they build their own specialty models based on, you know, all the training data they get from the users or they see what people really like to use, from a business perspective, it does make a lot of sense. They also get to raise a lot of money to justify doing that. Right? So this is sort of how I've been thinking about it. I don't know if you've used Windsurfers.

00:12:21.890 --> 00:12:37.865
I think it's SWE dash one. It was actually a decent experience. I didn't really get to use it as much, but it was very fast. It it it felt like the tabbing felt very fresh. It felt like it was a pleasantly good experience. So I do see cursor dash one or Windsurf

00:12:37.865 --> 00:12:41.225
version two coming out soon. So that's why I'm kind of like

00:12:41.625 --> 00:12:55.460
like and if you're like an AI builder, like, Cloud Code has an SDK coming out too where you can literally build on top of it. So it it seems to me they're they're definitely coming for everyone's lunch and Wait. Talk about that for a second. You said it has an SDK,

00:12:56.125 --> 00:13:30.195
which means people can build their own custom interfaces for it? So, yeah, instead of, like, you know, your your your tempos, your lovables, your bullets, your vibe code, like, instead of, like, building the tools for your agent and all that stuff, they're literally like, right now, it's terminal based, but their team said, if I'm not mistaken, yeah, the SDK currently supports command line usage, TypeScript, and Python SDK coming soon. So they're entering a world where they're telling the up and comers these guys will probably own always build their own stuff, but the up and comers, like, you focus on UX,

00:13:30.515 --> 00:13:36.835
we'll handle everything else for you. Right? So you can imagine yourself building a very clean seamless UX

00:13:37.075 --> 00:13:42.515
that literally runs on Claude code, and you end up having a model that competes with

00:13:43.030 --> 00:13:44.310
all these other guys.

00:13:44.630 --> 00:13:51.990
Right? So it it is an interesting I think there's gonna be some use cases for that that aren't, like, just broad coding

00:13:52.310 --> 00:13:56.145
things. Like, I feel like there you like, there's gonna be, like, microcoding

00:13:56.145 --> 00:14:00.065
tools for one specific niche that might be on the mobile device or

00:14:01.745 --> 00:14:12.190
I don't know. I've never thought about this before, but my mind is running with a bunch of different ideas. Because if you you could basically create a Claude code wrapper. And if you think Claude code is as good as Cursor,

00:14:12.350 --> 00:14:17.710
then, you know, we're gonna get, like, something that is as powerful as a Cursor wrapper. And that's terrifying.

00:14:17.710 --> 00:14:28.085
Like, you know, as someone who is we're building a a a a vibe coding startup, you know, it's it's terrifying to think that anyone can create a Claude Code rapper, but it's also very fun and interesting and maybe

00:14:29.525 --> 00:14:46.440
yeah. I don't I don't know. I don't even know what to say to that. I mean, what do you six months from now, what do you think that looks like? Like, what do you think people are gonna build on top of of Claude Code? You're right. What happens when everyone has access to the greatest agent, coding agent, and it can build on top of it? I I do think

00:14:47.640 --> 00:14:51.160
niche or niche, I never know which one is the correct pronunciation, specific,

00:14:51.240 --> 00:14:51.480
like,

00:14:52.175 --> 00:15:02.175
platforms are going to win. Like, for example, let's say you have a vibe coding tool that's specific to React. Right? Or you have a vibe coding tool specific

00:15:02.255 --> 00:15:14.870
to, like, WordPress or whatever. Right? I think, like, the niche tools are going to start to win because you can sort of, like I I I they haven't mentioned any fine tuning, but I can imagine, like, if you could fine tune the model to specific,

00:15:15.110 --> 00:15:15.830
like, types

00:15:16.310 --> 00:15:26.605
of programming or frameworks or stuff like that and you build, like, specialized user interface and you have content. Right? The model is definitely the model arena is definitely getting

00:15:27.725 --> 00:15:30.765
taken over by Cloud Code. And unless one of these guys

00:15:31.085 --> 00:15:37.800
and that's their goal, and that's why Cursor raised 900 freaking million dollars is because the only thing you can really build to withstand

00:15:37.960 --> 00:15:41.800
this behemoth is build another one. Right? So Build Anthropic.

00:15:41.800 --> 00:15:49.320
Yeah. That that's literally all you could do. Right? Because, like, at some point, like, the tool calling like, at first, I remember when GPT

00:15:49.560 --> 00:15:53.105
four just came out. We were, like, on 3.5 to four.

00:15:53.265 --> 00:15:58.625
Like, people were building, like, these open source. Like, I remember one of them was called, like, auto GPT,

00:15:58.865 --> 00:16:03.185
and the uniqueness was, like, being able to, like,

00:16:03.585 --> 00:16:12.920
have tools. Right? And tools were a big deal. Now tools have become normal, and then we have the MCP era, and, like like, this moat that these tools had

00:16:13.160 --> 00:16:14.760
is disappearing

00:16:14.760 --> 00:16:29.875
very, very slowly. And on top of that, they forked Versus Code. We can all fork Versus Code. Right? So the the the the big differentiator they can have is to build richer UX on top of what they have, which Cursor made a big bed background agents.

00:16:30.115 --> 00:16:33.795
They I don't know if you saw. They have their code reviewer, which is like a CodeRabbit alternative.

00:16:34.330 --> 00:16:51.175
Right? So they're building around the IDE, but then they're also going to build a model, which makes sense. But for And also they're also building a mobile app too. Like, Kurz was building a mobile app. I I I didn't even know that. So that's insane. Yeah. Yeah. Yeah. I follow their head designer on on X,

00:16:51.975 --> 00:17:05.150
and, yeah, they he's been posting it for a couple months now, some sneak peeks of what it will look like. So they're building a mobile app. So, yeah, I think yeah. I mean, it's it's probably time for Cursor to expand out into other niches and and try and figure

00:17:05.550 --> 00:17:08.110
out other ways to get some sort of edge. Because I do

00:17:08.750 --> 00:17:10.270
every single hike, like,

00:17:10.590 --> 00:17:22.215
at at our company, our back end developer was like, oh my god. Claude Code is so much better than Cursor right now. Like, he is, like, blown away, and he's just like, I literally tried Cursor and Windsurf

00:17:22.215 --> 00:17:33.440
for thirty minutes, and then Claude code just one prompted it. And, like, he'll just, like, enter a prompt and then just, like, walk away for seven or eight minutes and come back. The crazy part. Like and that's where, like, it actually

00:17:33.440 --> 00:17:48.985
feels agentic. I know we've been saying agentic this whole time, but Cloud Code is the one. And I sound like a shill. By the way, mind you, they're not paying me. I wish they were. But it genuinely it it's the first time I felt agentic, and it I'll I'll see it's the first time I felt like, yeah.

00:17:49.705 --> 00:18:04.360
Programmers who don't use these tools, yeah, like, you're you're you're definitely going to get, like, outshipped. Like, I know we've been saying this this whole time, but for the most part, I'm like, man, a really talented senior developer can, like, go and see. But Cloud Code is the first time where I'm like,

00:18:04.760 --> 00:18:08.360
yeah. Like, this this is real. Like, this is powerful.

00:18:09.000 --> 00:18:16.025
So one question I have for you is, like so Codex was recently released, like, in the last year by OpenAI.

00:18:16.025 --> 00:18:18.985
What is the difference between Codex and ClaudeCode?

00:18:18.985 --> 00:18:36.390
And then something like Devon, what about those three tools? Because it seems like they don't have kind of the standard interfaces like these three, uh, tools that you have here. What is the difference between these tools? I'll explain the difference in Codex and Cloud Code. So the underlying difference is Codex works on the cloud,

00:18:36.710 --> 00:18:40.710
and Cloud Code works on your local development, like, on on your computer.

00:18:40.870 --> 00:18:42.870
And there's pros and cons to both.

00:18:43.655 --> 00:18:46.055
I think the right bet

00:18:46.375 --> 00:18:47.415
is local

00:18:47.975 --> 00:18:49.175
just because

00:18:50.375 --> 00:19:13.245
like, and especially if you're building, like, serious applications and there's maybe certain things you need to install. Like, you're limited to what OpenAI allows you to install with Codex. Right? Versus with Cloud Code, I mean, on my computer, I can install whatever I want. I could start malware. Right? So, like, that's, I would say, the big differentiator and, obviously, the models that they use. I've used Codex before. Codex is pretty good.

00:19:13.805 --> 00:19:25.150
It's not Cloud Code, though. Like, I will just say this. In terms of, like, sheer power and output and speed, I definitely think Cloud Code takes it. But, again, when I look at Codex and them betting

00:19:25.710 --> 00:19:27.390
them making it on the cloud,

00:19:27.710 --> 00:19:50.460
it makes sense for, let's say, if you're a Vibe coder or you're not really comfortable. Like, I know many Vibe coders are comfortable with Cursor and Windsurf, but let's think even broader. Most people are not. Right? And if you can go from your, like, friendly chat assistant to being able to tell it to build something and they almost make you feel very intuitive and clean and simple, then Codex wins big with that market. Right? So

00:19:51.020 --> 00:19:53.660
I would say Right. You could almost make Codex

00:19:53.660 --> 00:19:55.900
a tool that ChatGPT

00:19:55.900 --> 00:20:02.940
could have access to in theory if it's in the cloud. Like, you could say, hey. Create an app, and you could literally just do this in ChatGPT,

00:20:02.940 --> 00:20:14.505
and then it just says, okay. Running that tool, and then maybe it sends it back to you, and it's just a web app link that you can open up right there. And you can look at it, you go back to ChatGPT and say, no. Change it. And, like, that is

00:20:15.065 --> 00:20:18.265
that's pretty big, and I don't I don't count that out for OpenAI.

00:20:18.730 --> 00:20:26.490
If there's one thing that OpenAI has proved to us over the year is that they have knocked the park out of consumer. Right? Perplexity

00:20:26.490 --> 00:21:15.830
held search down, and then ChadGBT came with their search massive. Right? Deep research came massive. Right? ImageGen came massive. Right? So I definitely wouldn't count them out in terms of, like, being able to build consumer products or things that consumers will wanna use. I just think in terms of, like, if I were to pick one, I would go with Cloud Code because Cloud Code just seems to and I think it's a it's a design choice that they went for the terminal because they don't really want to build the UX layer for you and all that stuff. Use whatever you're already using, and this is truly an agent that you just install in whatever you're using. Right? So I think it's a purposeful design choice, and that's what makes, in my opinion, Cloud Code better than Codex. It's more for me, for you, for someone who's a power user. Devin is different, though. I would put Devin

00:21:15.910 --> 00:21:17.030
in the same

00:21:17.510 --> 00:21:19.830
say, this is code review.

00:21:20.070 --> 00:21:22.550
Like, this is one spectrum of code review,

00:21:22.950 --> 00:21:24.390
and let's say this is another

00:21:25.245 --> 00:21:27.005
spectrum of, like, agent,

00:21:27.725 --> 00:21:29.565
I would put Devin

00:21:30.125 --> 00:21:32.205
somewhere around here.

00:21:32.765 --> 00:21:55.925
Right? Maybe maybe in the middle. Right? Somewhere, like, in this spectrum where I don't think and I haven't used Devon heavily, but based on the people I know at different companies who work at, they'll use Devon for speech feature specific team things. Right? So you already have an existing code base. Right? It's functional. It's in production. You have your Git workflows. You have a way of deploying to production. You have a staging environment.

00:21:56.245 --> 00:21:57.045
Devin

00:21:57.285 --> 00:22:00.725
integrates well with that. You can tell it what to do on Slack,

00:22:00.885 --> 00:22:03.205
and it will basically build you a feature.

00:22:03.445 --> 00:22:09.890
Right? Or it will review an existing feature that you've built. Right? You have then CodeRabbit.

00:22:09.890 --> 00:22:13.330
Right? And I'll maybe make this clearer for everyone to see.

00:22:13.650 --> 00:22:15.010
I would put

00:22:15.570 --> 00:22:17.010
let me just clear this up.

00:22:17.665 --> 00:22:19.825
I would put CodeRabbit here.

00:22:20.305 --> 00:22:21.825
And, by the way, I would

00:22:22.065 --> 00:22:49.945
100% use CodeRabbit. Like, their free tier is pretty good if your thing is open source. And if it's not, I think it's a couple dollars a month. Like, CodeRabbit is a great code review. So, like, whatever they do at I know they're not an agent, but what what at CodeRabbit, it is great. I can't tell you the amount of bugs I've caught using CodeRabbit. So excellent at what they do. Cursors So it it just just to clear things up. I've used CodeRabbit. I've actually have CodeRabbit. I used it, like, once. And to my knowledge,

00:22:50.185 --> 00:22:53.945
it's like when you ship a new feature and you upload it to GitHub,

00:22:54.025 --> 00:23:14.135
it will basically look at the changes you made and be like, hey. Like, this is good or like, no. You should change this. You could do this better, and it helps you it's like one layer to, like, catch mistakes or something like that. See, the thing is it's great because if you've noticed this, like, Riley, and I like I love your feedback, is whenever you start off a project with any model, really,

00:23:14.855 --> 00:23:28.200
it always starts strong. But as it continues to go on, like like, either performance goes out of whack or the model just started losing like, it's easy to start. It's hard to finish. Right? And there's a thing with, like, maybe the model has too much context.

00:23:28.280 --> 00:23:33.080
This is why they tell you create a new chat, start fresh, and all that stuff. With CodeRabbit,

00:23:33.080 --> 00:23:36.280
it gets your code all at once, and it just

00:23:37.115 --> 00:23:43.595
I think because it's not being repetitively used like the other, like, tools, it just is great at catching,

00:23:43.755 --> 00:23:45.515
like, either flaws

00:23:45.595 --> 00:23:46.875
or security

00:23:47.035 --> 00:23:49.035
mistakes or misoptimizations.

00:23:49.035 --> 00:23:58.570
Right? So I think So you're saying that, like, when you're using Cursor, because you need to manage your context and you can't speak into the same agent window for hours,

00:23:58.970 --> 00:24:08.605
that, like, it's gonna lose track of what you're doing in while you're using Cursor. But this is kind of a good thing. Like, once you end your full session, which might be 10 different agent

00:24:09.005 --> 00:24:09.965
chat windows,

00:24:10.045 --> 00:24:26.130
this will see it all at once and analyze it all at once so that you don't miss the big ones. Okay. So this is my technical theory. Right? It could be something else, but based off my experience, CodeRabbit, like, it's just been like and this is from, like I'm talking from also other developers. I'm using it at big projects,

00:24:26.450 --> 00:24:34.515
and it's just been great. And it's something simply you install, and it literally one thing that I love I don't know if I have a repo I can show, but it will

00:24:34.835 --> 00:24:37.235
give you a prompt to give the agent.

00:24:37.395 --> 00:24:43.155
Like, if you're using CloudCoder cursor, it will tell you, this is the bug that I found. Copy this prompt

00:24:43.395 --> 00:24:47.875
and paste it in whatever tool you're using, and it will fix it for you. Right? So

00:24:48.430 --> 00:24:49.950
I think this is one of those

00:24:50.270 --> 00:24:52.910
quick wins that if you haven't been using the program,

00:24:53.550 --> 00:24:56.590
they're not paying me. I wish they were. Hopefully, they do one day.

00:24:57.150 --> 00:25:02.190
It's it's a simple addition. And then for agent, we'll have Claude code

00:25:02.430 --> 00:25:04.725
on this side of spectrum.

00:25:05.045 --> 00:25:05.765
And then

00:25:06.485 --> 00:25:08.165
Devin is somewhere here

00:25:09.045 --> 00:25:09.845
where

00:25:10.165 --> 00:25:11.845
it's great at code review,

00:25:12.085 --> 00:25:13.685
but it can also build features.

00:25:13.925 --> 00:25:16.485
But I haven't heard any stories of people building

00:25:16.805 --> 00:25:17.525
Devin,

00:25:17.685 --> 00:25:19.510
like, something from scratch.

00:25:19.750 --> 00:25:20.630
And to

00:25:21.510 --> 00:25:22.950
the company's point,

00:25:23.190 --> 00:25:25.590
they called it, what, an AI software engineer.

00:25:25.830 --> 00:25:28.550
And more than likely for, like, big companies,

00:25:28.950 --> 00:25:36.945
like, a software engineer is not going to be building something from scratch. Right? They're probably gonna build and maintain features. Right? So if that's what they're building Devon for,

00:25:37.265 --> 00:25:44.385
that's perfect. So if you're a Vibe coder and your product hasn't hit product market fit and you haven't hired developers,

00:25:44.625 --> 00:25:50.060
I don't really think you need Devon. I think you get you something like a cursor windsurf

00:25:50.380 --> 00:25:57.180
plus Cloud Code if you wanna be fancy and CodeRabbit to review your stuff. I think you have a solid stack.

00:25:57.420 --> 00:26:18.035
K. That's yeah. You heard it here. And then so codex codex would be to the right of Devin Yeah. So codex left of Cloud Code. Yes. Something like somewhere right here. And mind you, not saying it's more of a code review, but I would just say even though I would align it similarly with car to code, I think Cloud Code is just more powerful simply for the fact that Codecs is cloud

00:26:18.710 --> 00:26:30.710
based. And that's the problem with cloud where, let's say, you wanna spin up a server. Let's say you wanna have multiple servers. Let's say you wanna you know, whatever that it is that you wanna do, there's a certain limitation you have with cloud versus with your own local machine,

00:26:31.375 --> 00:26:45.295
you can run and install whatever you want. Or even, like, security reasons. Right? Let's say, you know, I wanted to build a, you know, blockchain project. Why? I don't know, but let's say I do. Right? There there are certain security things and keys and stuff like that that I just wouldn't want on the cloud

00:26:45.690 --> 00:26:48.970
versus with Cloud Code, it's all on my machine my machine.

00:26:49.130 --> 00:26:56.490
So if I were to be launching into a long project, right, like you and I are gonna do in a future episode, where do you start? Right?

00:26:57.290 --> 00:27:03.655
I think that is a big question for VibeCoders. I think what, you know you know, Lovable, Tempo, Bolt,

00:27:04.455 --> 00:27:11.735
the whole nine yards, all those companies do a really good job of doing is kind of giving you a place to start from. You know? Like, if you just open up Cloud Code,

00:27:12.135 --> 00:27:19.510
it's hard to know what to do. I know you can open up Cursor, a fresh project on Cursor, and you can say, build all the files, and it will just do it. I guess, like,

00:27:19.990 --> 00:27:29.165
I guess in the next episode, when we when we build something from scratch using only AI and Cloud Code, how are we gonna start and, like, how are we gonna think about it?

00:27:29.485 --> 00:27:34.685
Here's the way I I I think of this. Right? And and this is stuff I've done for myself without AI.

00:27:35.965 --> 00:27:40.620
You realize that most projects have sort of the same base foundation.

00:27:40.860 --> 00:27:45.100
Meaning, you'll need a landing page. Right? Simple. You'll need authentication.

00:27:45.100 --> 00:28:01.975
Right? You'll want to authenticate the user. You want to store some user information. Right? You want the user to be able to log in with Google, email, whatever it is. Right? You'll want a database to store this information. Right? And here's one thing I wanna say. The one thing that actually frustrates me, Riley, is when I see vibe coders

00:28:02.215 --> 00:28:16.620
debate on which database platform makes more sense for you. It doesn't matter. Believe me. Until you hit PMF and it scales, then you can worry about it. But until then, bro, you use super base, you use convicts. By the way, I prefer convicts. But whatever it is that you use,

00:28:17.020 --> 00:28:29.425
doesn't matter. Right? It it it it doesn't matter. It's the midwit meme, you know, like, where, like, you know, the the people that mean, it's like it's like if if if the the database because I looked at the numbers. Right? It

00:28:30.465 --> 00:28:35.345
if when you're using Firebase versus if you're using Supabase or or InstantDB,

00:28:35.505 --> 00:28:37.105
you know, there's a lot of different tools you can use.

00:28:37.760 --> 00:28:48.160
It really doesn't become that big of an expense until you're really you're making us so much revenue that, like, hiring someone to come in and figure that out for you is gonna be

00:28:48.560 --> 00:29:12.090
very easy. Right? Yes. Especially if you're if you have the proper monetization in the app that you're creating. Right? Like and so, yeah, it doesn't really matter. And even I think the cursor windsurf debate is kind of not even worth it, really. Because at the end of the day, it's it's gonna come down to, like especially if you're vibe coding and you're not using AI, it's gonna come down to, like, how meticulous you are. Like and, like, it's gonna come down to grit, determination,

00:29:12.250 --> 00:29:14.090
and then discipline in, like, building

00:29:14.650 --> 00:29:32.515
actually building the one feature that you need to make a solid app, which is where I think a lot of people go wrong. They try and build a a ton of different features into an app, and then they just get lost when they should have and could have easily built that one feature. The I, like, I call it the halo feature that is at the center of their marketing,

00:29:33.350 --> 00:29:44.710
and they just could have marketed that one feature really well. And once you kind of have that revenue coming in, then you can worry about expanding it out and even hiring a team to come help you. We need a DB. We need auth,

00:29:44.790 --> 00:29:50.855
and we need to accept payments. And one thing I'll add, like, if there's one thing I would focus on is,

00:29:51.735 --> 00:29:52.375
like,

00:29:52.535 --> 00:29:55.255
there there's a term called DX developer experience.

00:29:55.335 --> 00:30:04.910
I would focus on tools that have that focus on ax, like AI experience. So, like, what do the AI models what are AI models good at? If you notice,

00:30:05.470 --> 00:30:07.870
TypeScript is a big thing with AI models.

00:30:08.510 --> 00:30:11.950
And the reason why the models will do good with TypeScript

00:30:11.950 --> 00:30:23.255
is because with TypeScript, there's a way to check if things are done right with types. Right? And the and this is getting technical, but the AI has a way to go through the code. And even though it might be hallucinating,

00:30:23.255 --> 00:30:31.095
it can check if it's done things right based on the rules of TypeScript. Right? So, for example, for me, I usually

00:30:31.255 --> 00:30:52.975
use the database provider or the back end provider convex because I realized that AI is really good with writing Convex code just because it's TypeScript. Right? At the end of the day, whatever you use, until you start to really, really scale, you won't feel it. Once you scale, you've made money, you're rich, congrats, hire some developers. You know? Send back the money to to the developers who who who built these tools.

00:30:53.215 --> 00:30:57.055
But I say that to say, if I were to build something using

00:30:57.135 --> 00:30:57.935
AI,

00:30:58.095 --> 00:31:03.400
I would first find a solid template. Now, Riley, I don't know if you have a template. I have two that I use

00:31:03.640 --> 00:31:05.960
on my GitHub. I have a React starter Back

00:31:06.760 --> 00:31:34.150
when we first started, yeah, we had templates, but they are out of date at this point because they go out of date relatively quickly. But I saw that Yeah. Yeah. Yeah. And I saw that yeah. You you messaged me that you have a template, and I've actually forked this before, and it works really well. And you're the first one who's started talking about payment templates, which is actually, like, one of the biggest things that, you know, people wanna make money with these tools. And so you've kind of built that in, and, obviously, there's keys you need to plug in. Right? If you have a Stripe integration,

00:31:34.310 --> 00:31:37.830
you need to plug in your Stripe account, and that can be learned over time.

00:31:38.230 --> 00:31:38.790
Yes.

00:31:39.030 --> 00:32:15.785
But, yeah, anyway, keep going. Yeah. So once like, I have two. One was with React star one's a React starter, one's Next starter. The only difference is this one uses Convex. This one uses a Postgres database. If you're not a developer, it doesn't matter. If you're a Vibe coder, I would start with the React starter kit, and we'll talk about this in the next video. I also have a video on it on my channel. It it's just easier to set up versus the next starter kit. There there's some, like, nuances that are a bit more difficult. So I would start with a starter kit. Get all my payments DB auth stuff figured out. And then what I would do using Cloud Code and this is this is this is key,

00:32:16.905 --> 00:32:39.835
Riley. I would tell Cloud Code to analyze the code base and create a markdown file explaining the flow of the app, like how it works, how do the payments work, what it used, etcetera, etcetera. The reason being is now I'm going to use that file as a base layer, a knowledge base for it to continue to build features. Right? So the plan is going to be find a template, download the template,

00:32:39.995 --> 00:33:00.840
set it up using Clockwork, tell me all the keys I need to set up, then build a markdown file telling me how the app works. And then what I'm going to do then is we're going to start building out features on top of this so we have our payments off and DB figured out. We just need to figure out the features and what we're going to build. And here's the cool part. I kid you not. Couple days,

00:33:01.160 --> 00:33:04.505
like for funsies, I would say 60%

00:33:04.505 --> 00:33:08.665
of the code is written by Cloud Code. I I wanted to build, like, my own

00:33:09.945 --> 00:33:10.825
my own

00:33:11.065 --> 00:33:12.025
CodeRabbit,

00:33:12.025 --> 00:33:16.105
but on the web. Right? And I log in with with GitHub.

00:33:17.000 --> 00:33:19.720
I'll click on analyze,

00:33:20.040 --> 00:33:22.760
and then this pulls all my public repos.

00:33:23.000 --> 00:33:29.320
I'll click analyze repo, and I just want to show you what you can do with Cloud Code. This is basically a full on editor.

00:33:29.720 --> 00:33:37.615
I I I did not I did not write the code for this. I I I Wait. Wait. You you use Claude code to make this tool called vibecheck.run.

00:33:37.615 --> 00:33:45.135
I saw I saw this on your Twitter, and I was just like, what is this? Is this a new cursor update? Like, what's going on here? This is and this is where it blew my mind.

00:33:46.020 --> 00:33:50.420
70 I can confidently say 70% of this code is written by

00:33:51.220 --> 00:34:01.665
quad code. And then I'll click, like, security analysis, and then I'll call the quad four o. I have an agent that runs through all the code and then will suggest

00:34:01.745 --> 00:34:02.625
changes.

00:34:03.025 --> 00:34:08.705
Right? And then these changes, I haven't worked on this feature yet, but I can push these changes to GitHub,

00:34:08.705 --> 00:34:20.820
create a pull request, and someone can merge that code change in. This took me two days. And when I say two days, I'm talking about two hours each day, and this was done through ClawdCode.

00:34:20.900 --> 00:34:26.020
Right? So in four hours, basically, in four hours, right, you could have done this in a four hour window.

00:34:26.260 --> 00:34:33.045
You created this. This is insane. Just so you could do the design alone. The design for the chat is fantastic.

00:34:33.285 --> 00:34:34.885
And here's the thing. And

00:34:35.125 --> 00:34:51.280
it this would've maybe two weeks. Like, I could have done this. Like, but the impressive part for me is not that it did something I could do. The impressive part is the time it saved. Right? And with Cloud Code, like, again, I skimmed through the code. I looked at it. I saw what it did. I checked everything.

00:34:51.440 --> 00:34:56.725
And, like, Riley, I I I did this in, like, four, five hours of just me relaxing

00:34:56.885 --> 00:35:02.085
and and and just chilling. And this is why I think, like, you find you a solid template.

00:35:02.245 --> 00:35:02.805
Right?

00:35:03.045 --> 00:35:05.125
You find your solid template. You,

00:35:05.680 --> 00:35:08.720
you know, get you cloud code to

00:35:09.120 --> 00:35:10.720
analyze the code base.

00:35:11.120 --> 00:35:15.120
By the way, don't judge me for my spelling. Cloud code base. And then you build features.

00:35:15.360 --> 00:35:29.495
I am telling you, you you're gonna have a great time. Will you get stuck at times? Of course. Will things get difficult? Of course. We don't need to hypothesize about it. We're gonna do it within the next week or two. We're gonna do this. So here's the here's the end of the episode, is we are

00:35:29.735 --> 00:35:30.615
the next time

00:35:31.175 --> 00:35:42.430
Mickey and I talk, we are going to build an app with Cloud Code starting from your template that's going to is it it's gonna have auth, it's gonna have payments Everything. All of it. It's gonna have everything,

00:35:42.590 --> 00:35:52.785
and we're gonna ship it live. And by the end of the episode, we're gonna have a link for people to go to to see it. Right? That's the goal. We're gonna make a whole episode. Make sure you spam, like, share, whatever it is.

00:35:53.585 --> 00:35:57.265
Hell, yeah. And then by the way, to everyone who's watching, I'm going to link

00:35:57.745 --> 00:36:50.665
Mickey's YouTube channel. I'm also gonna link your starter template, and then whatever else you want me to link, your Twitter, etcetera, etcetera. Thank you for coming on. This is actually I think it's really useful to put these tools next to each other to kind of get this mental market map, and I think it's actually useful to do with other categories as well. Like, I was trying to look at the difference between n eight n and Zapier and Make and try and figure out, like, what different tools are on the market. Like because you can't really use the tools unless you kind of know kind of it's it's always helpful to get that kind of perspective, and so I thank you. You're always, like, very good at explaining things. This is this was a lot of fun, so thanks for coming on. First of all, I appreciate the privilege. I appreciate the kind words. Thank you so much for having me all. And, yeah, like, I just like man, like, just have fun with the tools. Right? Like, a lot of them have free trials. Like, a lot of them are a couple bucks a month. Like, you know, just have fun. Like, don't take it too seriously until you have to. 100%.

00:36:50.665 --> 00:36:51.465
100%.

00:36:51.465 --> 00:36:54.585
Alright. I'll talk to you later. Alright. Bye, everyone. Thank you, Riley.