WEBVTT

00:00:00.000 --> 00:00:17.055
I gave the exact same prompt to Claude and Codex. Build me a full stack app from scratch, super base back end FireCore scraping deployed live on Vercel. Just here's what I want. Now go and build it. Then I had Gemini judge both code bases to decide who won. This is claw diverse codex head to head and the results are pretty hilarious.

00:00:17.135 --> 00:00:18.335
So let's get into it.

00:00:18.895 --> 00:00:21.935
Seriously though, this thing on the right is disgusting.

00:00:21.935 --> 00:00:24.015
I'm not sure how something

00:00:24.095 --> 00:00:29.240
can produce something with a font and spacing that looks like this in 2026.

00:00:29.240 --> 00:00:31.480
I mean, obviously, this one on the left isn't

00:00:31.800 --> 00:00:40.840
production ready or gonna win any awards for beauty, but it's certainly a lot better than what's on the right here. Alrighty. So I think the best way to do this test is to give them the exact same prompt,

00:00:41.160 --> 00:00:45.325
but give little room for interpretation so these things can think a little bit for themselves.

00:00:45.565 --> 00:01:13.005
There are a few more constraints though. One being that they're both gonna be running off of the same framework, which is inside our Claw dot m d, and this is my gotcha framework, which we'll get into shortly. And then inside our goals folder over here, we also have a build app dot m d, which has the Atlas framework inside, and that is a very specific framework for building apps. And we'll go into those while these things are building, but for now, just know that they are gonna be running in a completely fresh environment, each of them. They've both just initialized it based on their respective files,

00:01:13.245 --> 00:01:35.590
and they have the same prompt locked in. So we're gonna run this. I'm gonna leave most of the things up to interpretation, and we're gonna see the result that we get and just how well these things can function without any of my inputs. And, obviously, if they need my input, we will go and inspect that to see when and where they go wrong. So a quick overview of what we are building. We're gonna be building a competitive intelligence app called Rival. A user pastes in a company URL,

00:01:35.590 --> 00:01:38.230
and the app scrapes that company plus its top competitors,

00:01:38.615 --> 00:01:41.495
then displays a full competitive teardown dashboard.

00:01:41.735 --> 00:01:45.015
This kind of analysis would normally cost 10 k from a consultant.

00:01:45.255 --> 00:01:51.335
Then I just tell it that it needs to use my Atlas framework, so read goals forward slash build app. Inside there, we talk about Atlas.

00:01:51.690 --> 00:02:03.370
Then I give it some clues on the required stack that I want, and I'm mentioning this specifically because I wanna see how well it integrates with the MCP servers and does everything that it needs to do. So we're using Superbase for the database and authentication.

00:02:03.530 --> 00:02:05.975
We're gonna be doing web scraping with Firecrawl,

00:02:06.055 --> 00:02:15.095
and the deployment is straight into Vercel, and all of this is gonna be via MCP. The front end framework, this is where I'm giving them a little bit of freedom so they can pick whatever will produce the best result.

00:02:15.335 --> 00:02:18.615
We talk about some core features over here. I'm not gonna go too much into that.

00:02:19.260 --> 00:02:20.620
And we explicitly

00:02:20.620 --> 00:02:32.700
tell them that this thing needs to have AI in there, obviously, for the analysis and things like that. Then I give them just a very brief list of requirements. So the app must be deployed and live on Vercel when you're done. Superbase must handle all data persistence and authentication,

00:02:33.375 --> 00:02:36.895
and a few other things in here. One thing that I thought would be really important

00:02:37.135 --> 00:02:54.160
is that it must create its own instance of the Superbase project and the Vercel project. And I did that because I figured maybe one of these might finish a step sooner than the other one. And if it did that, then the other one might think, oh, hey, there's something in here. I can just go ahead and use this. And that might screw things up for this competition. So

00:02:54.560 --> 00:03:10.385
other than that, I think we're ready to go here. I was gonna measure time on this, but I think they're probably gonna run into a little mistake here and there. But we'll find out. So straight out of the batch here, we can see them in their planning process. This is obviously Codex 5.3 on the right, and on the left, we have Opus 4.6.

00:03:10.465 --> 00:03:15.745
So this thing is already asking us questions. Who's the primary end user for Rival in the first version?

00:03:16.065 --> 00:03:17.265
Let's just say founders.

00:03:18.670 --> 00:03:21.950
Which LLM source should power the analysis pipeline?

00:03:22.030 --> 00:03:23.470
So there we go. It asked.

00:03:23.790 --> 00:03:30.030
And, you know, I was talking to one of the AIs earlier, it was like, I bet OpenAI is going to recommend an OpenAI product.

00:03:30.190 --> 00:03:33.885
So, anyway, I'll go with this. I'm happy to go with its recommendations.

00:03:34.125 --> 00:03:37.485
For free tier operation, which run profile should be default?

00:03:37.725 --> 00:03:39.245
Let's just use balanced.

00:03:39.485 --> 00:03:52.070
How should competitor selection work in v one? It's recommending fully automatic. So again, I'll go with its recommendation here because I want this thing to supposedly tell me the best version of that this app that we can possibly get out of it. What authentication

00:03:52.070 --> 00:03:56.470
UX should we launch with? Email magic link as the recommended approach.

00:03:56.630 --> 00:03:57.910
Okay. We can try that.

00:03:58.310 --> 00:04:04.515
I would have maybe gone with Google auth depending on how big the app is gonna be. How should saved analysis be accessible?

00:04:04.595 --> 00:04:06.115
Private per user, definitely.

00:04:06.835 --> 00:04:12.995
So it's interesting that it's asked so many clarifying questions upfront, whereas Claude hasn't asked a single thing yet,

00:04:13.315 --> 00:04:14.915
but we'll see what its planning is like.

00:04:15.980 --> 00:04:18.460
What default usage cap should reinforce

00:04:18.940 --> 00:04:21.260
per user to stay in free tiers?

00:04:21.580 --> 00:04:23.180
K. If that's what you recommend,

00:04:23.420 --> 00:04:25.900
which visual direction should the dashboard follow?

00:04:26.140 --> 00:04:27.580
Bold editorial,

00:04:27.580 --> 00:04:30.620
enterprise clean, or playful startup? Let's go with playful.

00:04:32.505 --> 00:04:36.905
Which seed company should we use for smoke testing and first demo run?

00:04:38.025 --> 00:04:40.505
Fine. Okay. We'll use Linear. I don't really care.

00:04:41.625 --> 00:04:43.945
Wow. This thing is still just kind of planning.

00:04:45.170 --> 00:04:47.410
And it's looking at the Vercel documentation,

00:04:47.410 --> 00:04:51.730
is interesting. So I'm guessing it's trying to figure out how to actually deploy directly to Vercel.

00:04:52.050 --> 00:04:54.770
Which front end framework should I lock for implementation?

00:04:55.010 --> 00:04:56.450
I'll take your recommendation.

00:04:56.770 --> 00:04:58.690
How should we handle repeated analysis

00:04:58.935 --> 00:05:00.695
for the same target domain?

00:05:00.855 --> 00:05:02.775
Twenty four hour cache. That makes sense.

00:05:03.095 --> 00:05:06.535
Which OpenAI model profile should be the default analyzer?

00:05:06.615 --> 00:05:07.415
4.1

00:05:07.415 --> 00:05:08.215
mini?

00:05:08.695 --> 00:05:09.415
Really?

00:05:10.135 --> 00:05:11.495
I mean, maybe.

00:05:11.575 --> 00:05:14.055
But jeez, that's gonna be retired soon, surely.

00:05:15.230 --> 00:05:23.150
So I mean so I mean if it's recommending that again, I'm gonna choose what this thing recommends because otherwise it's kind of cheating. I don't wanna steer it too much.

00:05:23.630 --> 00:05:26.750
I guess it's not overly complex to just kind of sort through information,

00:05:27.355 --> 00:05:40.475
but one would think that if it's doing some kind of sentiment analysis or any deeper research, you might want a different model. But maybe it's trying to balance costs and I think if we actually read over here one final cost quality decision model tier for the analysis engine,

00:05:41.040 --> 00:05:45.680
Again, we'll see if it gets away with it. And coming in first place here, we have Codex,

00:05:46.080 --> 00:05:58.015
which has its full plan and actually asked a ton of questions upfront that were pretty good to get more of a user experience or workflow out of this thing. So let's expand the plan a little bit. I'm not gonna read through the whole thing

00:05:58.895 --> 00:06:12.810
because it's pretty in-depth. So v one implementation plan summary. Rival will be a Next. J s 15 app deployed on Vercel using Superbase for auth data real time and Superbase Edge function for orchestration. Awesome. So that's the choice I would have made.

00:06:14.170 --> 00:06:19.050
A user submits one URL. Rival auto discovers three direct competitors,

00:06:19.370 --> 00:06:23.555
configurable up to five, scrapes target and competitors with FireCrawl,

00:06:23.555 --> 00:06:29.475
runs structured analysis using GPT 4.1 mini, and and renders a live updating dashboard with history.

00:06:29.715 --> 00:06:30.595
That's awesome.

00:06:31.315 --> 00:06:38.170
So then it talks about the plan here, and you can see this is where it's starting to go through my Atlas framework, which we will go through very soon while these things are building.

00:06:38.490 --> 00:06:40.410
So the first phase is to architect,

00:06:40.570 --> 00:06:44.570
and it gives us our full app brief problem, users, success, constraints.

00:06:45.210 --> 00:07:01.815
Although, this isn't as in-depth as most of my builds are with Claude, but we'll see what it does this time around. Next up, we have Trace and this is where it goes through the data schema and figures out all the moving parts from that end. So you can see here, Superbase and then it lists all the different schemas in here. We have an integrations map,

00:07:02.775 --> 00:07:04.535
Superbase, FireCrawl, OpenAI.

00:07:05.140 --> 00:07:22.595
Okay. That all looks pretty good and we've got some free tier controls in here. Then next up, we have link and this is where we check that everything is actually working before we go ahead and build so we can see all of the moving parts over here. And we'll go and check this before it just randomly builds the entire system and then has to figure this out after it's built something.

00:07:23.155 --> 00:07:27.075
Then we have the assemble part and this should be layer by layer. So hopefully, it's done that.

00:07:27.395 --> 00:07:37.240
Initialize app structure, implement super base schema, implement edge orchestration. Yes. So it's doing it piece by piece rather than just trying to one shot something super amazing. It's gonna take our layered approach.

00:07:37.960 --> 00:07:41.160
And then finally, stress test. So it's gonna be doing functional tests

00:07:41.320 --> 00:07:42.920
and integration tests

00:07:43.000 --> 00:07:54.365
and any edge case tests. Great. Then we have validate, which is the security and the correctness. So we obviously have to bake this in because if we don't have any security protocols on there, if we don't do any security checks as a part of our app development,

00:07:54.685 --> 00:08:07.220
it's not a very good app, is it? Now, realistically, there's only so many tests that this thing can run and obviously, if we have real users, they'll be doing user acceptance testing and a whole much deeper form of testing. But this thing just does what it can upfront.

00:08:07.380 --> 00:08:15.235
Then last up, monitor. We have some operational telemetry. So persist stage timings and failure reasons in analysis events, track run duration.

00:08:15.475 --> 00:08:35.000
There are lots of parts that could have taken here to monitor this stuff because both Superbase and Vercel have some form of observability in them. We could also have used Helicone or something like that for the OpenAI agent and various other things we could have done locally as well, but it didn't suggest any of those. To be honest, we weren't really that specific on that front. It's just pulling this out of my framework because my Atlas framework

00:08:35.160 --> 00:08:49.485
basically cements this in that we need to do this sort of thing. So that's a pretty cool plan. We're gonna go ahead and implement this very soon. I just wanna have a look at the questions that Claude is asking me over here. And it looks like the first question this thing is asking me is if I have an API key for Anthropic

00:08:49.485 --> 00:09:08.600
because it needs it for the edge function. So it looks like both of them have chosen a super base edge function for the AI, which would probably be the right move. Edge functions are probably the best way to do this kind of thing. So I'm gonna say, yes, I have one, and I'll provide this off screen to the AI when it asks for it. It did also offer GPT four o as a secondary option.

00:09:09.475 --> 00:09:34.720
But more interestingly here is that this thing didn't ask me a single question apart from the API key. So it's taken the entire build into its own hands without asking any clarifying questions and things like that, which is a very interesting approach. Let's have a look at what its plan looks like. So problem getting competitive intelligence today requires hours of manual research across competitor websites or paying 10 k for a consultant. Rival automates this. Paste the URL, get a full competitive teardown in under ninety seconds.

00:09:35.040 --> 00:09:37.040
And then it goes straight into Atlas, so architect

00:09:37.675 --> 00:09:44.395
talk about the problem, the users, the access, and constraints. So again, fairly similar to OpenAI's if we come and have a look over here.

00:09:45.195 --> 00:09:49.275
Same for bullet points, so it's kind of aligned with mine. Put this in a table for trace,

00:09:49.740 --> 00:09:57.420
which is just easier to read, suppose, for us. So layer the choice that it made and why it made it. This one didn't do that. So that's also really interesting.

00:09:58.380 --> 00:10:04.220
So the front end, it chose NextJS 14 plus and it's chose that because it's native Versal Deploy.

00:10:05.215 --> 00:10:10.095
And styling, Tailwind CSS. Okay. Okay. Okay. The AI analysis.

00:10:10.415 --> 00:10:12.735
Anthropic Claude Haiku 4.5,

00:10:12.895 --> 00:10:16.015
cheapest Claude model. Yeah. Sure. That makes sense, and it's fast.

00:10:16.950 --> 00:10:20.230
And I'm pretty sure this thing is gonna outperform what OpenAI recommended,

00:10:20.310 --> 00:10:21.750
but we'll find out very soon.

00:10:22.470 --> 00:10:25.990
Long running pipeline, super base edge function, 150

00:10:25.990 --> 00:10:35.915
wait time versus the sales ten second limit. I wonder if it actually did research into that to make sure that that's accurate. Key architecture decision, super base edge functions for orchestration.

00:10:36.235 --> 00:10:40.955
And I chose that because the Vercel hobby plan has a ten second function time out. Far too short.

00:10:41.670 --> 00:10:47.430
Okay. Cool. And then we got to database schema. So migration one, two, and three, and four.

00:10:48.150 --> 00:10:49.510
Integrations map.

00:10:50.390 --> 00:10:53.190
All looking good. Atlas phase link connection validation,

00:10:54.135 --> 00:10:58.215
assembling everything. So it printed a very different plan for assemble.

00:10:58.775 --> 00:11:00.615
If we look at assemble over here,

00:11:01.095 --> 00:11:33.415
just the structure is very different. This one didn't give us a breakdown of everything. Cool. So I'm gonna approve both of these plans and I'm also gonna add the API key off screen so that this thing doesn't run into any problems. And while those things are building, I'm gonna give you guys a very quick overview of the system. I've gone massively in-depth in other videos, so you can check that out if you want long form. I'm just gonna give you some context as to how the system handbook works, how our build app framework works, and then I'll show you how to set all of this up manually and how you can integrate in your environment so you can then just go away and get this thing to build anything that you want, really. So firstly, we have claude.md.

00:11:33.415 --> 00:11:46.800
You can change this to any filename you want. Both Claude and Codex are using this right now. If you're switching to Codex, all you do is change this to agents dot m d and it will do the exact same thing. So we have the gotcha framework here. It's a six layer architecture

00:11:47.360 --> 00:12:04.715
and this just governs everything inside Versus Code or anti gravity, whatever you're using. We have our goals and this is what needs to happen. So you can see here one of our goals that comes standard is to build an app. We'll get into those very shortly. Then we have the orchestration layer, which is the manager, and this is the thing that governs everything inside our environment.

00:12:05.280 --> 00:12:20.480
Then we have tools, and these are deterministic scripts that will do the job each time in the exact same way. The reason that we have this three letter acronym over here is because AI is probabilistic. So one day you might get something like this out of it, and the other day it might do something completely different based on the exact same input.

00:12:21.005 --> 00:12:32.445
So we stop that happening by giving it deterministic tools, and those two things work together in order to have the probabilistic nature of AI, but we have that determinism for business logic and systems that we're building.

00:12:33.730 --> 00:12:57.425
Cool. But we can take it much further than that. So then we have context because, obviously, you have a business or system or anything that you're building, it has context behind it, and the AI can use that context as part of whatever it is that it's building. So for instance, in inside my folder, I have everything to do with my life, my business, my content, and this environment here can just pull that out for me. It's really useful if you're building specific things where you need that context.

00:12:57.745 --> 00:13:06.700
Then we have prompts, and all that is is literally a prompts folder with reusable prompts that we stash there so we can use them for regression testing outside of a Python script,

00:13:06.940 --> 00:13:35.630
or we can use them in other systems in a whole bunch of different scripts and things like that. And then finally, we have arguments which are just variables, and they go in at runtime for specific things that change during runtime as opposed to having to change them from a script every single time we run them. So again, just know that this thing kind of governs our entire system and there's a whole bunch more that goes in here. We've got guardrails baked in and again, you can check out the full length video if you really wanna understand how this file works. Just know that it governs everything within our environment and that's the important part here.

00:13:35.950 --> 00:13:42.830
But then more importantly, when we're trying to build, we don't just wanna yellow our way into every single build. We wanna give it some structure, some best practices structure.

00:13:43.305 --> 00:13:51.385
And that's where Atlas comes in. So this is a five step process for when I'm building MVPs, and it becomes a seven step process if I'm building something closer to production.

00:13:51.865 --> 00:14:01.870
So the first step that we have here is architect. This is where we define the problem, the users, what success looks like. Because in context engineering, you need to have an understanding of what good looks like.

00:14:02.190 --> 00:14:23.575
If you don't have that, the system couldn't possibly know where to take you. This is arguably the most important step in the entire process because the more specific you are and the more planning you do upfront, the less work you're gonna have to do later on. Next up, we have trace, and this is where we define the data schema, the integrations map, your tech stack, and things like that. You would wanna feed as many ideas of the tools that you're gonna be using upfront.

00:14:23.840 --> 00:14:35.600
Again, And you can work with the AI beforehand to figure this stuff out. It's not like I walk in there for every single build that I'm doing, like, yes, I want to do this. I know exactly what I'm doing. No. You can think with the AI to fill all of this stuff in for you.

00:14:36.385 --> 00:14:56.070
Then next up, we have link, and that's just where we validate all the connections. And then we have assemble, and that's where we build our layered architecture step by step. Start with basic functionality and you move upward rather than trying to just shoot for the moon and have to fix a bunch of shit later on. Then we have stress test. It's very important to test the functionality, the error handling before we actually do anything.

00:14:56.390 --> 00:15:28.950
And finally, bake in security and some monitoring because those are vital for any real app out there. So again, that's just these files at a very high level. Check out the deep dive on the screen now. But But what I'm gonna do is you can grab these files in the description below and you can use them as your framework to build whatever you want. I'm gonna show you how to set those up. Cool. So all you would have to do to get started is drag these into Versus Code. This is all part of the full tutorial video that I just mentioned a few seconds ago. So I'm not gonna go fully into the weeds with this because again, that will show you how to do all of this. So I got my little Claw chat here and I'm just gonna say initialize this environment.

00:15:29.270 --> 00:15:38.585
And then what it's gonna do is it's gonna go ahead and read its Claw dot m d. It's gonna understand what the system handbook is and it should build out our structure exactly as we want it. So I'm gonna bypass permissions.

00:15:38.905 --> 00:15:53.810
We can see there it's already gotten started and you can see how it's starting to build out our tools. We've got our context folder. We've got our goals, which was already there, but it should add a manifest very, very soon. We've also baked in memory, so that was an addition that I made in that other video that I keep referencing,

00:15:54.210 --> 00:16:05.795
where we're trying to bake in memory of this just like Claude Bot had, but but this is obviously way more secure. While our other environment is initializing, let's check-in on our builds. So this thing is just over halfway. That's Claude.

00:16:06.035 --> 00:16:10.675
And over here, it doesn't really have a to do list. It just seems to be kind of quirking away in the background.

00:16:11.155 --> 00:16:21.000
We'll flip back to this in a second. And there we go. This thing's done a few seconds later. So you can see it's now set up our entire environment over here with our memory system and the manifests that we need for our apps.

00:16:22.600 --> 00:16:33.075
And look at that. One of them is finished and one of them is asking me a question and kind of failed but not really a failure, I guess, at the end of the day. So we'll go into what it built very shortly.

00:16:33.635 --> 00:16:43.075
But on the right hand side here, Superbase reports that creating a new project will cost $0 monthly. Please confirm you understand that and want me to proceed. I absolutely do.

00:16:44.070 --> 00:16:45.030
Continue.

00:16:45.110 --> 00:16:46.070
That's fine.

00:16:46.710 --> 00:16:48.390
$0 $0.

00:16:49.350 --> 00:16:55.590
Meanwhile, we can come over here and have a look. Rival is live what was built. The database, super base project rival with three tables.

00:16:56.165 --> 00:17:01.605
The back end, Superbase Edge function analyze company v two with full scrape and AI pipeline.

00:17:01.765 --> 00:17:03.765
The front end, Next. JS16

00:17:03.765 --> 00:17:04.885
plus Tailwind

00:17:05.285 --> 00:17:08.085
authentication is email password via Superbase.

00:17:08.730 --> 00:17:12.890
So Codex recommended the password list via a magic link,

00:17:13.210 --> 00:17:16.250
and this one recommended the traditional email password.

00:17:16.890 --> 00:17:20.010
At the very least, I would have added Gmail. I mean, come on.

00:17:20.410 --> 00:17:23.370
Deployment versus production, yes, it's exactly what we told it.

00:17:25.115 --> 00:17:34.075
And we'll come and open this up very, very soon. I just wanna see what this thing is saying on the right hand side here. Super base project creation is blocked by your account's free tier limit.

00:17:34.555 --> 00:17:35.275
Ah.

00:17:35.675 --> 00:17:36.475
That's interesting.

00:17:36.990 --> 00:17:40.750
So it's blocked because I've reached the max active free projects, is two.

00:17:41.230 --> 00:17:48.190
Alrighty. I just deleted one, so let's tell it. I just deleted one of the older projects, so you can go ahead and create it now.

00:17:48.990 --> 00:17:53.315
And we'll let that thing go off to the races. We'll come back and reveal these when the other one is done.

00:17:53.635 --> 00:17:56.515
Okay. And this thing is still chugging along here a little bit.

00:17:56.915 --> 00:18:08.740
Having a bit of trouble with super bass or doing something very, very in-depth compared to whatever Claude did. Another question. Do you want me to deploy rival to Vercel production now with required super based environment variables?

00:18:08.900 --> 00:18:10.180
I absolutely do.

00:18:10.340 --> 00:18:14.740
I wonder if this thing's gonna run at all on the left because it literally asked me nothing.

00:18:15.060 --> 00:18:16.900
So so far, already off the bat,

00:18:17.220 --> 00:18:17.940
I think

00:18:18.260 --> 00:18:23.955
Codex is superior in terms of asking user clarification and functionality upfront.

00:18:24.275 --> 00:18:28.515
Claude just kind of says, I know what's best and nobody cares about what the user wants.

00:18:29.075 --> 00:18:33.795
And then literally built the entire thing without asking me a single question about it apart from the API key.

00:18:34.970 --> 00:18:38.490
So it'll be interesting to see if its intelligence slash arrogance

00:18:38.570 --> 00:18:41.530
actually amounts to something over Codec's ability.

00:18:42.010 --> 00:18:44.330
Alrighty. We got some more things going on here.

00:18:44.810 --> 00:18:46.810
Production deploy started successfully.

00:18:46.810 --> 00:18:47.850
The URL's been issued.

00:18:48.465 --> 00:18:52.945
Vercel blocked production because NextJS fifteen point two point four is vulnerable.

00:18:53.105 --> 00:19:00.625
I'm upgrading to a patched NextJS line now. I wonder if this thing ran into the same version or if it automatically just went for NextJS

00:19:00.785 --> 00:19:01.425
16.

00:19:02.625 --> 00:19:07.380
That'd be interesting to figure out. But this thing seems to be healing itself and just carrying on with life.

00:19:08.020 --> 00:19:17.060
And more questions here. Do you want me to install Next JS sixteen point one point six? Oh my god. Yes. Just do it. And this thing is still chugging away. Ask me another question

00:19:17.775 --> 00:19:19.055
about security,

00:19:19.295 --> 00:19:26.415
whether it should expose the public URL. And another one, do you want me to verify the deployed API health thing? Yes. Just do everything.

00:19:26.975 --> 00:19:33.455
So this thing is very needy. But I think they're doing it to bake in more security for people who are less security conscious.

00:19:33.695 --> 00:19:41.270
So they might stop, think about this, and then maybe ask this thing or, you know, Google it to see what the consequences might be. I'm not sure if that's intention.

00:19:41.430 --> 00:19:44.550
Do you want me to verify the deployed run analysis? Yes. Do it.

00:19:45.270 --> 00:19:46.630
So that's quite frustrating,

00:19:46.965 --> 00:19:50.565
and I'm sure you can bypass it. You probably just need to update your settings

00:19:51.125 --> 00:19:59.125
somewhere inside here to completely disable any form of it asking you anything. Claude just meant that a lot easier by having a bypass permissions button over here.

00:19:59.770 --> 00:20:03.610
Oh my god. I think it's finished and I have aged significantly.

00:20:04.010 --> 00:20:07.930
Implemented end to end rival v one in this workspace and deployed it live.

00:20:08.730 --> 00:20:10.250
Cloud resources created.

00:20:11.370 --> 00:20:29.680
Okay. That's fantastic. So it looks like it built a bunch of stuff. Cool. So let's get this thing going then. We need one manual step. Both of them recognized it, but Codex had a difference here in the sense that it needs to add the super base edge function secrets in the project. Wants me to do that manually. This thing does not mention that. That's really interesting.

00:20:30.080 --> 00:20:34.720
But we can easily go and do that. And then both of them identified that we need to change the URL

00:20:35.040 --> 00:20:43.905
inside the auth. So I'm gonna do that off screen and then we'll jump back when it's ready. Okay. So Codex is still stuck in some kind of test verification

00:20:43.905 --> 00:20:44.545
loop.

00:20:44.865 --> 00:20:51.585
And as you can see, I'm just asking it politely why it's taking so long to do all of this stuff when Claude finished ages ago.

00:20:51.745 --> 00:20:56.145
It's now literally been two hours, and I'm still sitting here. I've been sitting silently

00:20:56.600 --> 00:21:05.960
waiting, hoping, and even praying, which goes against every part of my being. And we still haven't finished. I think it is broken. I think I've officially broken codex

00:21:06.120 --> 00:21:07.560
or run out of something.

00:21:08.040 --> 00:21:09.160
We still have context,

00:21:09.495 --> 00:21:16.455
quite a few tokens, but obviously, it can compact it if it ran into any kind of problem. I definitely still have cred. I checked that as well.

00:21:16.935 --> 00:21:22.535
So I'm really not sure what's going on at this point. But we're gonna check Vercel and let's have a look at what's going on inside

00:21:22.775 --> 00:21:24.135
our console. So

00:21:25.540 --> 00:21:28.260
Codex Rival, this is the Codex one up.

00:21:28.420 --> 00:21:29.540
It's rebuilding.

00:21:30.980 --> 00:21:32.980
Okay. So at least it is doing something.

00:21:33.940 --> 00:21:35.860
It's been building for seven seconds.

00:21:37.060 --> 00:21:41.625
So let's leave that one alone then. Okay. So Codex finally deployed it,

00:21:41.945 --> 00:21:44.025
but as you can see, there's an error.

00:21:44.345 --> 00:21:45.785
So that's not a good start.

00:21:46.665 --> 00:21:48.905
So this is actually terrible

00:21:49.065 --> 00:22:04.950
and we're gonna reveal which one is which very soon. So what we did for testing here is we have a third party involved because it's not just me. I'm not just gonna judge this aesthetically and presume that the back end is all hunky dory. What we're actually doing here is we're gonna judge this, we're gonna test this ourselves,

00:22:05.455 --> 00:22:23.860
But then I've also been using Gemini Pro three, and it's busy looking at the back end and at our front end code to do an assessment for both of those to give its own judgment on who it thinks is superior at what. So So let's do some basic tests and then we're gonna flip over to Gemini and see what it thinks of this thing's building capabilities.

00:22:24.260 --> 00:22:27.380
Seriously though, this thing on the right is disgusting.

00:22:27.380 --> 00:22:29.380
I'm not sure how something

00:22:29.460 --> 00:22:34.625
can produce something with a font and spacing that looks like this in 2026.

00:22:34.625 --> 00:22:36.945
I mean, obviously, this one on the left isn't

00:22:37.185 --> 00:22:42.145
production ready or gonna win any awards for beauty but it's certainly a lot better than what's on the right here.

00:22:42.625 --> 00:22:45.505
But I'm not gonna be that negative. Let's test functionality.

00:22:45.505 --> 00:22:46.945
So why don't we just do

00:22:47.345 --> 00:22:50.910
let's do clickup.clickup.com.

00:22:51.630 --> 00:22:53.950
Maybe if I could spell clickup.com.

00:22:54.190 --> 00:22:56.430
And then we'll just copy and paste this in here.

00:22:57.790 --> 00:23:01.710
So if either of these work first shot, I will be massively impressed.

00:23:02.345 --> 00:23:04.345
Because remember this was a one shot.

00:23:04.585 --> 00:23:12.745
Ah, okay. So this one wants me to sign up first before we're allowed to do anything. So I've signed up to the left hand one. Let's see if this has the same functionality. Run teardown.

00:23:13.625 --> 00:23:20.630
Sign in. Yes. So this won't let me do anything without signing in either. But this one has a magic link. So let's test the magic link and see if that works.

00:23:21.110 --> 00:23:24.150
Oh, you're losing points there, buddy. Doesn't work.

00:23:24.550 --> 00:23:26.870
Let's try refreshing the page and trying again.

00:23:28.965 --> 00:23:30.805
Nope. That's still not working.

00:23:31.125 --> 00:23:35.685
So straight off the bat, it failed immediately for the second time trying to do its thing

00:23:36.645 --> 00:23:52.060
which obviously isn't very good. So I'm going to fix this off screen and we'll try log in again. Okay. So it fixed that problem. I'm gonna enter my email in here and sign up off screen again. Okay. So one of them is running on the right and let's do the other one on the left. If these work, honestly, I'll be amazed.

00:23:52.700 --> 00:23:58.575
If I had to guess which one is gonna work, I would probably say none of them but maybe the one on the left hand side.

00:23:59.215 --> 00:24:00.415
Let's find out.

00:24:00.815 --> 00:24:03.855
There we go. We already failed on the right hand side.

00:24:04.255 --> 00:24:07.775
Edge function returned a non two x x status code.

00:24:08.330 --> 00:24:15.290
So this is super generic. It's just saying a non two x x. So it's looking like 200 is normally okay.

00:24:15.610 --> 00:24:31.685
It doesn't even give us a proper error code. So we're gonna have to go and scrape through the back to see if we can find anything here. And at this point, it looks like they're both broken just as I thought neither of them would work. So left hand side, we've got Claude which at least looks better and in the right hand side, have Codex which took

00:24:31.925 --> 00:24:34.485
hours and hours to do absolutely nothing

00:24:34.805 --> 00:24:35.765
terrible and broken.

00:24:36.370 --> 00:24:43.010
But we'll fix this in the background. For now, let's flick on over to our anti gravity and have a look at what this thing when thinks who's the winner.

00:24:43.570 --> 00:24:46.450
So first up, I reviewed the security and architecture

00:24:46.450 --> 00:24:49.090
for the back end because obviously, that's completely separate.

00:24:49.575 --> 00:24:53.895
So the clear winner here was actually Codex, if you can believe that.

00:24:54.535 --> 00:25:07.750
So the winner is Rival twenty twenty six, which is our Codex build. The newer project demonstrates a significantly more mature and robust security architecture. It implements comprehensive row level security, which is pretty basic and also very important.

00:25:07.990 --> 00:25:13.990
CRUD operations includes immutable audit logging and utilizes performance optimized authorization patterns.

00:25:14.230 --> 00:25:15.750
Lots and lots of words.

00:25:16.630 --> 00:25:21.715
You can see this is a tie. This is probably the most basic thing that your database should have enabled for security.

00:25:21.955 --> 00:25:24.355
So it was a tie. No brainer there.

00:25:24.835 --> 00:25:28.115
But this thing says that Codex is the clear winner here,

00:25:29.075 --> 00:25:32.915
which is interesting, but I suppose also not surprising because it took significantly

00:25:32.915 --> 00:25:34.675
longer at planning,

00:25:35.170 --> 00:25:37.810
but also implementing and it asked me a bunch of questions.

00:25:38.130 --> 00:25:39.650
So I'm sure that that influenced

00:25:39.810 --> 00:25:47.730
the security as well. And if you look at the update for what Codex 5.3 was supposed to bring is that it is more robust in its security.

00:25:48.050 --> 00:25:50.050
So that's clearly showing on the back end here.

00:25:51.055 --> 00:25:52.735
However, on the front end,

00:25:53.135 --> 00:25:55.935
for the code base itself, things are a little bit different

00:25:56.175 --> 00:25:58.735
and you can see here the clear winner is Claude.

00:25:59.215 --> 00:25:59.775
So

00:26:00.575 --> 00:26:03.535
completeness, full auth flow, login sign up confirm,

00:26:03.695 --> 00:26:05.455
minimalist callback only.

00:26:06.110 --> 00:26:14.270
So it probably picked up that it didn't do the login first the first time when I scanned this because I scanned this before I had actually logged into either app

00:26:14.830 --> 00:26:20.670
and that's probably why I picked up that there was a problem. CoreLogic dedicated API analyze endpoint placeholder.

00:26:21.535 --> 00:26:22.255
Okay.

00:26:22.655 --> 00:26:24.415
Not sure what this thing is analyzing.

00:26:25.615 --> 00:26:26.895
UI complexity,

00:26:26.895 --> 00:26:29.295
10 static pages, six architecture,

00:26:29.615 --> 00:26:30.495
237

00:26:30.495 --> 00:26:31.295
dependencies,

00:26:31.295 --> 00:26:32.495
139.

00:26:33.135 --> 00:26:33.935
Stability,

00:26:33.935 --> 00:26:35.535
how is it getting this?

00:26:35.775 --> 00:26:36.575
That's interesting.

00:26:39.430 --> 00:26:44.630
Alright. So it's literally just from a build. I wouldn't say that that is a stability of the platform then.

00:26:45.270 --> 00:27:00.615
So perhaps we'll rerun the code based part of this after we get everything working. So I've got to say that Claude is still the clear winner for me like for every single thing that you do, Claude code is just better than Codex. I hate Codex's user journey. I hate its user experience.

00:27:01.015 --> 00:27:05.575
It is vastly slow in comparison to Claude. Even if I use Sonnet,

00:27:06.100 --> 00:27:08.340
Codex is slower than Sonnet.

00:27:08.580 --> 00:27:13.700
I don't like the way it thinks. It does random things very slowly as well.

00:27:14.260 --> 00:27:20.100
Everything in here is completely predictable. It lets me know what it's doing, and I realize you can click on here and see anyway.

00:27:20.895 --> 00:27:27.535
But something to do with the user journey of how Claude approaches it is just far superior to me,

00:27:28.015 --> 00:27:29.935
specifically in its troubleshooting as well.

00:27:30.415 --> 00:27:41.820
There is no way you could convince me to ever use Codecs over Claude. I just wouldn't do it. I realized both of them failed, but realistically, you have to look at this. We built a somewhat complex,

00:27:41.820 --> 00:27:43.340
obviously not really, but

00:27:43.740 --> 00:27:45.100
for a one shot,

00:27:45.580 --> 00:27:48.060
every single part of these functionalities to work flawlessly.

00:27:48.615 --> 00:27:52.135
It's not really realistic. There's always gonna be a little bit of iterating

00:27:52.215 --> 00:28:03.175
especially when I gave it such vague instructions. If I gave it better planning upfront and I was very specific in what I wanted and how it worked and all of that, we would have got a much better one shot from the beginning. But the point of this test

00:28:03.660 --> 00:28:15.180
was to prove this kind of thing. I wanted to see what it could get away with if it had to think a little bit for itself and then just go and build. And this is where we got to. And Claude is definitely far superior on the sense of the aesthetics,

00:28:15.735 --> 00:28:19.975
but also troubleshooting. I mean, look how much faster this thing is working than this thing on the right.

00:28:20.375 --> 00:28:28.535
And Codex on the right had much more of a head start because I started troubleshooting this thing long before I started Claude. But we have to take into account anti gravity's critique

00:28:28.910 --> 00:28:50.625
with the back end. Obviously, I'm not going go through the database and the back end and verify any of this information. I'm going take it for its word. It used the MCP server. It logged straight into the projects and read both of them. So we have to take it on its word for that much. But this one I will rerun if we get these apps working. Otherwise, again, we're just going to take it on its word because from what we can see on the front end,

00:28:51.265 --> 00:29:07.880
it's vastly superior. It's a win for Claude. We don't really need Gemini's perspective on this. We can clearly see who the winner is on this side. In terms of front end functionality, yes, they are both broken and I'm gonna give this about five to ten more minutes to see if I can get either of them working properly.

00:29:08.120 --> 00:29:17.565
And if that doesn't work, then I'm just gonna cut this off because I've been recording for two hours and thirty four minutes and for the most part that is because of Codex and its literal

00:29:17.805 --> 00:29:21.885
slow time. Claude finished maybe an hour at least

00:29:22.365 --> 00:29:28.525
before Codex did. It just seemed to keep iterating through things and MCP servers and doing

00:29:28.810 --> 00:29:29.850
random things.

00:29:31.290 --> 00:29:37.050
And again, they both had the exact same prompt to start, so it's not like I'm doing something funny in the background here.

00:29:37.450 --> 00:29:39.690
So Claude has fixed its problem allegedly.

00:29:39.690 --> 00:29:52.565
The edge function analyze company was deployed with verified JWT true, which requires a valid JWT in the authorization header. Cool. So did you fix it? No Vercel redeployment needed. The fix was entirely on Superbase side.

00:29:53.205 --> 00:29:56.645
Alrighty. You should now be able to go here and log in.

00:29:57.205 --> 00:29:57.605
Okay.

00:30:03.350 --> 00:30:11.030
K. And we're off on a second test here using our competitive analysis while this thing is still trying to troubleshoot itself and just figure out life.

00:30:11.270 --> 00:30:14.390
So while this thing's running, I'm gonna see if we can trail any logs

00:30:14.565 --> 00:30:19.045
to see if this thing's actually running or if it can pick something up in real time if it is failing.

00:30:19.765 --> 00:30:27.045
It's working. Your new ClickUp analysis is actively running and is currently on the analyzing step. The final stage, here's the progress so far.

00:30:28.550 --> 00:30:33.990
Okay. So then it's a front end problem where it's just not updating these little bubbles. But let's see if it actually finishes.

00:30:34.230 --> 00:30:35.590
Codecs

00:30:35.590 --> 00:30:37.510
is still trying to fix itself.

00:30:38.710 --> 00:30:41.190
Let's have a look at our app.

00:30:41.525 --> 00:30:44.405
Have you finished? Nope. Still hasn't finished.

00:30:44.805 --> 00:30:48.405
What we can do is we can actually check FireCool to see if it's done anything.

00:30:48.965 --> 00:30:55.605
So the time is wrong here. It thinks I'm in a different time zone. But for the most part, it looks like it has been scraping competitors

00:30:56.140 --> 00:30:59.980
of ClickUp. So we got Monday, we got Notion, we've got Atlassian,

00:31:01.180 --> 00:31:02.140
Asana.

00:31:02.540 --> 00:31:05.900
So one of these has been scraping. I'm guessing it's obviously the Claude.

00:31:06.220 --> 00:31:07.260
Let's have a look.

00:31:07.820 --> 00:31:10.165
So they had scraped discovered competitors.

00:31:10.165 --> 00:31:16.325
Yes. Okay. So the Claude is winning so far. AI analysis failed to pass competitive analysis results.

00:31:16.645 --> 00:31:20.565
The problem is max tokens. It's too small for a full competitive analysis.

00:31:21.300 --> 00:31:23.940
We can see here on the right that codecs is just

00:31:24.740 --> 00:31:26.020
it's just getting nowhere.

00:31:26.180 --> 00:31:34.020
So, from a troubleshooting perspective, I mean, we're doing it live right now. This thing's already fixed two problems in the time that this thing's trying to figure out where left and right is.

00:31:34.675 --> 00:31:36.195
It's massively different.

00:31:36.915 --> 00:31:41.715
And this is what I'm talking about with the user journey. I don't know how to explain it or articulate it,

00:31:42.355 --> 00:31:46.675
but even the even when this thing is wrong, it's so confident about it that it just makes me love it.

00:31:47.910 --> 00:31:53.350
So that's why I've been using it. That's the why the only thing I pay for for the max plan is Claude.

00:31:53.750 --> 00:31:55.670
I promise I'm not sponsored by them.

00:31:56.070 --> 00:32:00.870
So edge function v four deployed three three key fixes, max tokens 8,000,

00:32:00.870 --> 00:32:11.695
try pass JSON, cool. Now try again. I'll do that. Okay. So it had a retry button, which I also like. I mean, I didn't ask for that functionality, but it had it and I clicked it. And here we go.

00:32:12.335 --> 00:32:16.415
So we can flick back over to fire crawl to see if this thing's doing anything.

00:32:17.560 --> 00:32:19.080
Let me refresh this.

00:32:20.600 --> 00:32:24.680
We should see some in progress unless this thing is so super fast that it's just

00:32:25.640 --> 00:32:31.400
that it's just running through it or maybe it will use a cached version. Who knows what it's doing? Nothing's been kicked off yet.

00:32:31.975 --> 00:32:37.335
Let's ask it to trail the logs and see if it can do that. On the right hand side here, we're still exploring

00:32:37.735 --> 00:32:38.535
life.

00:32:39.095 --> 00:32:40.855
Six files, two surges.

00:32:40.855 --> 00:32:43.095
What is what is it even doing?

00:32:46.070 --> 00:32:51.430
Pulling super base edge function logs now to see the exact HTTP status body behind.

00:32:52.470 --> 00:32:53.990
I mean, is mental

00:32:54.150 --> 00:32:56.390
that it takes us long to figure out something.

00:32:56.870 --> 00:32:58.150
Claude destroys it.

00:32:58.835 --> 00:33:02.595
It's running great on v four. Here's the live status edge function of post.

00:33:02.835 --> 00:33:07.315
No more four o ones. So if we come back here, do we see anything? Not yet.

00:33:08.115 --> 00:33:11.715
I hit the retry button. So how are you saying that one is still pending?

00:33:12.440 --> 00:33:17.560
Or what is happening with the one that was running now? The newest one completed successfully.

00:33:18.680 --> 00:33:27.640
Fully worked, but it didn't populate the front. I mean, let's try click back. So, this completed. Ah, there we go. So, there's still some kind of functionality problems here

00:33:28.345 --> 00:33:32.105
in terms of how it presents it to the front end because we had to refresh the page.

00:33:32.425 --> 00:33:37.865
But realistically, we're gonna go through this. I'm just gonna stop bothering with codecs now because this is ridiculous.

00:33:38.185 --> 00:33:43.545
We fixed four errors in the time it took to do whatever the hell is going on here. But it found a bug.

00:33:44.470 --> 00:33:49.030
So it's gonna go fix that bug while it fixes that bug. I'm just gonna come and take a look at this.

00:33:50.790 --> 00:33:52.150
Let's zoom in.

00:33:52.710 --> 00:34:00.195
ClickUp positions itself as an all in one replacement for a 100 plus fragmented tools, competing against established players like Asana, Monday, and Notion.

00:34:00.435 --> 00:34:05.155
So this is pretty cool. I mean, we have features, pricing, positioning, content, gaps,

00:34:05.635 --> 00:34:13.020
and each of them can be clicked on. So obviously, a UI perspective, I mean, it's not entirely beautiful, but it's functional and

00:34:13.340 --> 00:34:18.540
it far surpasses anything that Codex even bothered to get out. Even though it failed, this thing fixed itself.

00:34:18.700 --> 00:34:23.820
Codex couldn't even do that properly. So it's just a very clear table. We've got our pricing over here.

00:34:25.665 --> 00:34:34.545
ClickUp freemium, all of these have freemium, I guess, but so that's just a feature comparison and then a pricing comparison for the features that you'd get. Positioning,

00:34:35.505 --> 00:34:38.145
ClickUp software to replace all software strengths,

00:34:38.310 --> 00:34:41.830
clear memorable headline with strong consolidation value prop.

00:34:41.990 --> 00:34:46.470
Weaknesses, replace all software. Claim is hyperbolic and unsubstantiated.

00:34:46.630 --> 00:34:48.870
May alienate enterprise buyers.

00:34:49.110 --> 00:34:49.750
Alrighty.

00:34:50.230 --> 00:34:50.870
Asana,

00:34:51.095 --> 00:34:55.415
strengths Fortune 100 adoption creates significant credibility moat.

00:34:55.655 --> 00:34:56.455
Weaknesses,

00:34:56.455 --> 00:34:59.335
Fortune 100 focused may alienate SMBs.

00:35:01.175 --> 00:35:03.495
Notion is your AI everything app.

00:35:03.655 --> 00:35:09.520
I'm pretty sure a lot of apps are saying that now. Monday, outpace everyone with the best AI work platform.

00:35:10.160 --> 00:35:13.040
So these are pretty straightforward, but again, at least it's working.

00:35:13.360 --> 00:35:23.725
Content. So this is the content that each of them are putting out and you can see it's giving us a little summary over here. ClickUp and monday.com minimize blog investment instead embedding content through our product and demo pages.

00:35:23.885 --> 00:35:28.525
Asana and Notion prioritize blog and case studies for thought leadership and SEO.

00:35:29.405 --> 00:35:30.605
So again, that's pretty cool.

00:35:31.430 --> 00:35:37.510
And then gaps. So this is where it would be helpful to us if we were trying to get into this market ourselves and see how we could be different.

00:35:37.910 --> 00:35:40.950
Enterprise customer success and managed services narrative.

00:35:41.270 --> 00:35:50.535
ClickUp mentions CSM and managed services on enterprise tier, but no competitors prominently feature customer success stories, case studies or ROI testimonials.

00:35:50.855 --> 00:35:57.335
Asana leads here with Fortune 100 adoption claim, but lacks published customer success stories or quantified outcomes.

00:35:58.010 --> 00:36:02.810
So you can see this is pretty good. I mean, it it did exactly what we wanted based on our original prompt.

00:36:03.050 --> 00:36:04.330
In every single way,

00:36:04.810 --> 00:36:10.970
Claw delivered on everything that we wanted. Yes. It didn't work the first time around, but again, we were not hyper specific

00:36:11.050 --> 00:36:22.255
and it's normal to have some kind of bug with minor iterations. These were not giant things. It solved it in less than five minutes and if I didn't spend time waiting for Codex, we would have been done ages ago.

00:36:23.135 --> 00:36:25.855
But now let's flip back over to Codex and see what's cracking.

00:36:26.760 --> 00:36:31.240
So the error was coming from super base returning four zero one on run analysis.

00:36:31.320 --> 00:36:35.800
So a very similar error to what we had on the other side except this thing took

00:36:36.040 --> 00:36:38.360
nineteen minutes and ten seconds

00:36:38.760 --> 00:36:47.865
to find the exact same problem that this thing took I don't know what was it like thirty seconds to a minute to figure out and then solve in under four minutes.

00:36:48.265 --> 00:37:02.370
So that is an immense difference in in time. I mean, it's it's ridiculous. I have no idea why the hell this thing takes so long. And for those of you wondering, yes, I am running this on the highest possible thought patterns that Codex has as well. So it's benchmark

00:37:03.010 --> 00:37:04.290
to benchmark here.

00:37:07.010 --> 00:37:10.885
And then we rerun it again and it says invalid JWT.

00:37:10.965 --> 00:37:19.285
And that's it. That is how I'm going to end this video. I'm not wasting any more time on codecs. There is no way you could convince me or pay me to use codecs.

00:37:19.365 --> 00:37:23.285
And that's not me being a dick. That is because the user experience I have had

00:37:23.760 --> 00:37:26.320
from the get go of this has been horrible.

00:37:26.640 --> 00:37:33.680
Both from the way that it works inside Versus Code all the way through to building and iterating with this thing, it is not a pleasant journey.

00:37:33.920 --> 00:37:43.095
It might have much better security on the back end, but if it can't get the front end functioning at all, then we have massive disconnect between security and functionality.

00:37:43.095 --> 00:37:52.940
And realistically, when you're building anything, you need to have that perfect little triad or at least put it in favor without destroying the ability to use the thing properly.

00:37:53.100 --> 00:38:04.220
And again, that's why I think Claude is massively superior here because they seem to be getting better and better at that balance while also riding this massive hype train that they have at the moment without destroying their own product.

00:38:04.815 --> 00:38:13.935
So I think they're doing amazing things and Gemini backs up obviously their front end. Sure. Maybe their back end security and things like that could be better. But realistically,

00:38:14.095 --> 00:38:16.175
if you're gonna be sending this to kind of thing to clients,

00:38:16.800 --> 00:38:26.080
you would be doing pen testing, you would be getting proper developers involved to make sure that this is getting done the way that it should be so that you're selling a real product. This was just MVP,

00:38:26.320 --> 00:38:28.800
and Codex failed to build an MVP,

00:38:29.040 --> 00:38:36.665
which the majority of platforms out there can do for $20 or less. And Claude Sonnet could easily have done if I didn't bother using the Opus model here.

00:38:36.985 --> 00:38:55.665
Anyway, that's where I'm gonna leave this thing. Let me know any comments you have down below. I will get back to all of them. Otherwise, check out the videos on the screen right now. You can also look at my community where we've just launched the vibe coding course as well as a whole consulting path based on everything I've done over the last twelve years. So if that resonates with you, feel free to check it out. Otherwise, I'll see you guys in the next one. Thanks for watching.