Claude Opus 4.6 vs GPT-5.3 Codex: Live Build, Clear Winner

I sit down with Morgan Linton, Cofounder/CTO of Bold Metrics, to break down the same-day release of Claude Opus 4.6 and GPT-5.3 Codex. We walk through exactly how to set up Opus 4.6 in Claude Code, explore the philosophical split between autonomous agent teams and interactive pair-programming, and then put both models to the test by having each one build a Polymarket competitor from scratch, live and unscripted. By the end, you'll know how to configure each model, when to reach for one over the other, and what happened when we let them race head-to-head. Timestamps 00:00 – Intro 03:26 – Setting Up Opus 4.6 in Claude Code 05:16 – Enabling Agent Teams 08:32 – The Philosophical Divergence between Codex and Opus 11:11 – Core Feature Comparison (Context Window, Benchmarks, Agentic Behavior) 15:27 – Live Demo Setup: Polymarket Build Prompt Design 18:26 – Race Begins 21:02 – Best Model for Vibe Coders 22:12 – Codex Finishes in Under 4 Minutes 26:38 – Opus Agents Still Running, Token Usage Climbing 31:41 – Testing and Reviewing the Codex Build 40:25 – Opus Build Completes, First Look at Results 42:47 – Opus Final Build Reveal 44:22 – Side-by-Side Comparison: Opus Takes This Round 45:40 – Final Takeaways and Recommendations Key Points Opus 4.6 and GPT-5.3 Codex dropped within 18 minutes of each other and represent two fundamentally different engineering philosophies — autonomous agents vs. interactive collaboration. To use Opus 4.6 properly, you must update Claude Code to version 2.1.32+, set the model in settings.json, and explicitly enable the experimental Agent Teams feature. Opus 4.6's standout feature is multi-agent orchestration: you can spin up parallel agents for research, architecture, UX, and testing — all working simultaneously. GPT-5.3 Codex's standout feature is mid-task steering: you can interrupt, redirect, and course-correct the model while it's actively building. In the live head-to-head, Codex finished a Polymarket competitor in under 4 minutes; Opus took significantly longer but produced a more polished UI, richer feature set, and 96 tests vs. Codex's 10. Agent teams multiply token usage substantially — a single Opus build can consume 150,000–250,000 tokens across all agents. The #1 tool to find startup ideas/trends - https://www.ideabrowser.com LCA helps Fortune 500s and fast-growing startups build their future - from Warner Music to Fortnite to Dropbox. We turn 'what if' into reality with AI, apps, and next-gen products https://latecheckout.agency/ The Vibe Marketer - Resources for people into vibe marketing/marketing with AI: https://www.thevibemarketer.com/ FIND ME ON SOCIAL X/Twitter: https://twitter.com/gregisenberg Instagram: https://instagram.com/gregisenberg/ LinkedIn: https://www.linkedin.com/in/gisenberg/ Morgan Linton X/Twitter: https://x.com/morganlinton Bold Metrics: https://boldmetrics.com Personal Website: https://linton.ai

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:00Greg Isenberg?

  Today's a massive day because Anthropic just dropped Opus 4.6 and OpenAI answered with GPT 5.3 Codex. But what is the better model and how do you get started? And what are some tips and tricks to get the most out of them? Well, this episode is all about that. This is for the technical person who's trying to get the most out of these models, who don't just want hot takes, who want tactical sauce for getting the most out of these models. This episode of the pod is with my dear friend Morgan Linton. Morgan is one of the best engineers I know. He was an executive at Sonos. He's invested in a lot of AI companies and he's building an AI company of his own. He's one of my first calls when I'm like, hey, which model's better?

• 0:45Greg Isenberg?

  So we put the models head to head and there's a winner at the end. We rebuild Polymarket, a multi-billion dollar app, but we use these models. So which is the better one? You'll find out by watching this episode. But you'll also learn to become a better AI developer because you'll have these tips and tricks in your back pocket.

• 1:12Greg Isenberg?

  I'm with one of my favorite people, Morgan Linton. You might not know him, but he is just, you know, just an incredible developer, founder, entrepreneur, investor. He does it all. But today, what I needed him to help me understand is Opus 4.6 just came out. GPT 5.3 Codex just came out. Morgan, help me understand by the end of this episode, what are people going to get out of this?

• 1:38Speaker 01

  Yeah. Well, Greg, thanks for having me. Super exciting day. It's moving fast today. Opus 4.6 came out and then Sam Altman put together a quick tweet. I want to say like maybe 18 minutes later announcing GPT 5.3 Codex. And me, I think everybody else has been jumping on it, playing around, figuring out the differences. You know, all the little neat new settings that there are in each of these. By the end of this, you're going to know first how to make sure that you are running Opus 4.6 and all of the little details you can change. Settings dot JSON file to use some of the cool features in Opus 4.6, especially agent teams, which is probably the feature I'm the most excited about.

• 2:19Speaker 01

  You'll also understand why you might use one versus the other because they both kind of tackle different engineering methodologies. And then hopefully you'll see some cool stuff as we build some demos together that I've put together that I haven't tried myself. So I'll be trying just live with you. So we'll, we'll see how that goes.

• 2:37Greg Isenberg?

  Cool. I think one of them is we're going to try to recreate Polymarket.

• 2:41Speaker 01

  Yes.

• 2:42Greg Isenberg?

  And see which model performs best.

• 2:46Speaker 01

  They're going to have both. They're going to do a head to head to try to each build their own version of Polymarket.

• 2:51Greg Isenberg?

  So by the end of this episode, you will have a pretty good understanding of how to use the models, when to use the models, how to get started. Morgan, let's get into it.

• 3:01Speaker 01

  Cool. Right on. All right. So I took some notes and essentially, you know, with 5.3 Codex, I'll be showing that in the desktop app on Mac because they're super excited about that. I'm excited about it. I think if, if OpenAI was wanting a demo to be done the right way, they would want me to do it in their app. Whereas with Opus 4.6, I would say the Anthropic team would want me to do it in the CLI. And so there's a few different configuration settings that you do want to make sure that you get right when you're using Opus 4.6 or trying to use Opus 4.6 today, tomorrow, whenever it is that you're jumping in to use it. I've seen a lot of people online today on Twitter saying, it's weird. I'm having a problem. Like

• 3:45Speaker 01

  I'm supposed to be agent teams, but I don't see them or how do I know what version I'm running? So I thought, let's start by just giving everybody a level playing field to know, okay, I want to be able to use Claude code with Opus 4.6. How do I make sure I'm doing that and doing that correctly? So here's kind of the initial to do's that everyone should have on their list. Just do an NPM update. See if that does the trick. If that doesn't, and you're running an older version, then run Claude update. But you, you should see like as of right now, it's 2.1.32. If you see one dot something, you're running an old version. And then what you want to do is go into your settings dot JSON and I'll

• 4:27Speaker 01

  just show this here. So if you just do like CD, uh, tilde slash, uh, Claude.

• 4:37Greg Isenberg?

  So I bet that there's people who are running the old model, they don't even realize it.

• 4:42Speaker 01

  I have no idea.

• 4:44Greg Isenberg?

  Bad result.

• 4:45Speaker 01

  Yeah, yeah. So I mean, make sure you go in here, cd slash dot Claude. Here's your settings dot JSON.

• 4:53Speaker 01

  If you view this, here's essentially what you should see. Now, it's okay, it can be model. If you want to like really be specific about it, you can put in Claude dash opish dash four dash six, that will lock it in. But because four six is the newest model, you can also just put in model and just opus and that'll work. The key thing that you want to do is, in my opinion, the coolest feature that they added with four six is agent teams. I'm super excited to demo that with you. You have to make sure to turn that on because it is an experimental feature. And that's probably the biggest confusion I'm seeing people have today with opus four six is that they are running opus four six.

• 5:35Speaker 01

  They keep hearing about agent teams and they're giving it prompts like build a team of agents, do this and this. And it's not quite doing it. And that's because you do have to enable this. So you do have to add and end this Claude code experimental agent teams and then set it equal to one. Okay. Nothing too crazy. Once you do that, that will make all that possible. So with that in mind, you're pretty ready to go there. Then you can just run Claude in the terminal and you're good. For people that are using the API, the one thing I did want to point out is there's a pretty cool new addition, which is called adaptive thinking. Also, just to be clear, because I'm seeing confusion on this too.

• 6:18Speaker 01

  This is in the API, this is not in Claude code itself. But adaptive thinking, just to show it here. You're able to essentially pick the level of effort that you would like the model to use. This is only going to work in four six, by the way, if you want to use like an effort level of max. And so here's kind of the different levels. So with max, Claude always thinks with no constraints on thinking depth. It's opus four six only. So requests using max on other models are going to return an error. So. If you're calling the API and you set the effort level to max and you get an error, then you're probably not using OPUS 4.6. But here's the example where you can see if I'm calling the API,

• 6:59Speaker 01

  I set the model to Claude OPUS 4.6 and then here's where I can set the effort. And this is another thing, if you're using existing API code, you may have the model of OPUS 4.5 and now you adjust the effort to max. It gives you an error. All you need to do is just bump the version and you're good. But this is kind of a neat thing they've added to the API with 4.6. It's worth mentioning. And then kind of the last, the last thing I would say is just if you want to use split panes for agents, so if you want agents to show up in different panes and you're using something like Warp, just make sure to install TMX. You can do this with brew install TMX.

• 7:41Speaker 01

  And then if you do that, it's going to default to auto, which usually means in process, which means in that same terminal window you have, the agents are going to be working all together. If you want it to split pane, then you just need to update that setting in the settings.json on the split panes. I'm not going to go into super details on that, but those are just like, I think, good housekeeping to start with for anyone using OPUS. But don't worry about it. Really, all that anybody needs to do, especially if you don't even want to use teams, agent teams, is just make sure you're updated using the newest version and that the model is OPUS and you'll be using OPUS 4.6.

• 8:21Greg Isenberg?

  Cool.

• 8:22Speaker 01

  So that's that. Before I get into kind of the differences between OPUS 4.6 and Codex, I thought I would actually read this because this was posted on Hacker News four hours ago and I was reading it and I was thinking, that's like the best way to explain it. So I'm just gonna, I'm just gonna read this little section here because I think they do such a good job with it. This person's saying, "What's interesting to me is that GPT-5 and OPUS 4.6 are diverging philosophically and really in the same way that actual engineers and orgs have diverged philosophically." I think this really nails it. "With Codex 5.3, the framing is an interactive collaborator. You steer it mid-execution, stay in the loop, course correct as it works.

• 9:06Speaker 01

  With OPUS 4.6, the emphasis is the opposite, a more autonomous, agentic, thoughtful system that plans deeply, runs longer and asks less of the human." That feels like a reflection of a real split in how people think LLM-based coding should work. Some want tight human loop control, others want to delegate whole chunks of work and review the results. And I honestly, I think that says it beautifully, and I think that that nails the differences. And also hopefully, you know, there's all, everybody wants to pick a winner where it's like, oh, no, no. Opus 4.6 is better, Codex. It's different. It depends on what your methodology is.

• 9:44Speaker 01

  And I think what we're seeing now, not just with vibe coding, but also with like overall like AI-powered engineering is how do you want to work with agentic coding? Do you want to have a totally autonomous experience where you're sending agents out to do work? Or do you want to work with an LLM like another teammate and pair program with the LLM? And that's where you're now seeing a divergence where I think you're going to see a lot of teams using both. Because Codex really is your collaborator. And what they've added with 5.3 is like really good, like mid-execution steering.

• 10:23Speaker 01

  Whereas with Opus 4.6, it's probably the best of the best now, being able to say, I want to spin up three or four agents. I want them to go do stuff. Hey, and don't bug me. I want to trust they're going to do good stuff. And it's able to deliver.

• 10:38Greg Isenberg?

  So are you saying that there in some ways, it's just a preference? Like depending on how, you know, there's no right or wrong, basically, you know, you're not wrong to be an Opus person or, you know, it just like might feel, yeah, it's just a preference.

• 10:57Speaker 01

  Yeah, well, you might be both, right?

• 10:59Greg Isenberg?

  That's true.

• 11:00Speaker 01

  It might turn out that you're both. That's why, like, not to disappoint people here, but we're not going to end this with me saying, and so the winner is, it's like, well, depends on what you want to do. Everyone has a different methodology for it. So I'll dive in and try to, try to make this part fast because I know the fun part is probably us going in and playing around with both of these and having them do a head-to-head and try to build a competitor to Polly Market in however much time we. But I'll just, I'll just start kind of going into these at a high level, just so for anyone who wants to know, like, what are the core differences? Why is this so interesting? Just going to what that is. So with, with OPUS 4.6, much bigger context window. So you have a million token context window here.

• 11:47Speaker 01

  Very strong coherence over entire documents and repos. Designed for, you know, like load the whole universe and reason over it. 5.3, they talk about large context, but it's not a headline feature. And I actually went back and forth with it to get it to actually give me a number. And the number is around 200,000 tokens, which is not that impressive. That's smaller than I was thinking it would be. But that's okay. It's optimized, you know, for, for progressive execution rather than total recall. So that's why that's not as important. And, you know, optimized for deciding like what to keep in working memory. So high level, what that means is Claude is better when the task is understand everything first and then decide.

• 12:32Speaker 01

  GPT 5.3 Codex is probably better when the task is decide fast, act, iterate, more of that, you know, pair of programming, you know, mid-task change behavior. For, for coding benchmarks, you know, OPUS 4.6 is really good at code-based comprehension. Refactors with like architectural sensitivity, explaining why a system behaves a certain way. And then, you know, a little less tendency of this like Yolo write code, right? Which is, I think, something everybody wants. So, you know, that's good for everybody, but especially for vibe coders that are getting started and

• 13:14Speaker 01

  they may not be able to identify hallucinations. OPUS 4.6 is definitely going to perform better there. But then for teams, you know, building in large code bases like, like, like me and my team are doing. That's also really important. So kind of a win for everyone there. 5.3 Codex did win on SWB Bench Pro, Terminal Bench. Overall, it's like scored better on coding benchmark.

• 13:36Speaker 01

  So probably better end-to-end app generation. And you know, Claude's kind of like senior reviewer staff engineer. GPT five three, probably like your, your founding engineer, right? Uh, agentic behavior, uh, Opus four, six. This is the key one, right? Is like the multi-agent orchestration. That's, that's probably like the bleeding edge feature in four, six. And then with, uh, with, with, uh, five, three codex, really like task-driven autonomy. Build, test, modify without being asked. But then this task steering, you can watch it. You can go in and it's like your buddy's coding and you can say, oh, wait, wait, man. Wait, why are you doing this? And you can stop it and it'll go, okay. And then you can restart.

• 14:20Speaker 01

  You can really fix things in line. Much harder to do that with, with Opus. With Opus, you'll kind of be stopping it and then starting somewhat fresh. But it has a pretty big context window, so it, so it knows what it did. But you know, Claude's really asking like, should we do this? GPT, GPT five three is like, how fast can I ship this? Right?

• 14:40Greg Isenberg?

  It's really, I mean, it's so cool because it almost feels like they're different people. You know what I mean? Like they have different styles.

• 14:48Speaker 01

  Yes, totally. Yeah, it's a good look at it. It's like a different, it's like a different personality type, right? And then, yeah, uh, failure, failure modes, uh, you know, Claude four, six, it might overanalyze. It's got a much bigger context window. Uh, can hesitate when, when requirements are ambiguous. And then it can stop short on a full execution. Uh, five, three codex, uh, could be overconfident, can lock in a flawed assumption early, but you can steer it back in the right direction if that, if that happens. So that's kind of a high level overview on the, on the two.

• 15:25Greg Isenberg?

  Cool. That's helpful.

• 15:27Speaker 01

  Yeah. So should we, should we just dive in? I haven't tested any of this. So this is now I have like zero canned demos because I thought it'd be more fun just to try something together and see what happens. So should we, should we try it?

• 15:41Greg Isenberg?

  Yeah.

• 15:41Speaker 01

  Okay, so let's see, I'm gonna, I'm gonna start with Opus. And I've got these prompts preloaded. So I'm giving different prompts, just like, I think you said it really well. It's like you're talking to different people. And so, you know, when I'm talking to Opus, I can tell Opus, build me a team. And here's what I want each member of the team to do. When I'm talking to Codex, I can't really tell it to build me a team, but I can tell it to think about stuff. So the prompt that I'm going to give to Opus is, build a competitive polymarket. Create an agent team to explore this from different angles. One teammate on technical architecture, one on understanding polymarket and the ins

• 16:22Speaker 01

  and outs of prediction markets, one on UX, and one that just works on building really good tests to make sure everything works. For Codex, I'm going to give it a little different prompt, but very similar. So still build a competitive polymarket, but now think deeply about technical architecture. Understanding polymarket and the ins and outs of prediction markets. Good, clean UX. Make sure it builds really good tests to make sure everything works. And to be fair, I'm going to try to paste these in around the same time.

• 16:52Greg Isenberg?

  You're a fair guy, Morgan.

• 16:54Speaker 01

  I'm trying to, I'm trying to keep it fair here, right? It's the only way to do it. Like I said, no, no winners or losers. It's, it's just about, just about letting everybody have a fair shot to play the game.

• 17:05Greg Isenberg?

  Yeah.

• 17:06Speaker 01

  All right, so let's see, I'm going to make different directories. So I'll do, uh, let's just call this Opus45 polymarket competitor. All right, so let's fire up Claude in here. By the way, if you want to check when you're running, uh, just to like really make sure that you're in a good place with the model, if you type slash model, I can see here, right? Claude Opus46, right?

• 17:36Speaker 01

  So I'm good there. I'm going to take this prompt, copy it, make sure this is all copied in correctly.

• 17:45Speaker 01

  Okay, got that. I'm not gonna hit Enter yet. I'm making this totally fair. I don't want anyone at Anthropic or OpenAI to get upset with me. So I want to be in good terms with both of them.

• 17:55Greg Isenberg?

  Totally. Smart guy.

• 17:57Speaker 01

  Let's see. Oh, wait. Actually, you know what? I do want to create a new folder for this.

• 18:02Greg Isenberg?

  But we are keeping it real. We're being objective. Neither myself or Morgan are affiliated with either... Well, actually, I don't know about you. I'm not affiliated with Anthropic or OpenAI.

• 18:12Speaker 01

  Nope. Nope. I love them both equally. How about that?

• 18:17Greg Isenberg?

  Yeah.

• 18:19Speaker 01

  Okay. And I'm gonna try to start as close to the same time as I can. Enter. Go. All right, they're going. Off to the races.

• 18:30Greg Isenberg?

  So what do you think's gonna happen?

• 18:33Speaker 01

  That's a great question. Well, I know right now, because I told Opus45 to build using different teammates, it's gonna do that. So you can see here, it says, "I'll build a polymark competitor by launching parallel research agents first, then synthesizing their findings to a comprehensive implementation plan and codebase." This is brand, brand new, right? Like, if I did this with Opus yesterday, wouldn't be possible. That's kind of the difference here is that the way that Codex is working is the way things have kind of always worked, right? So if you see, this is like the individual person, right? It's

• 19:15Speaker 01

  not saying, "Okay, I'm gonna launch all these different agents and compare what they say." It's like, "Okay, I'm gonna inspect the workspace." This is your, you know, really detail-oriented, really senior-like founding engineer, like that example gave, right? Whereas over here, you can see, it's already launched these agents, and now it wants to do web searches, and I'm gonna let it do that. So multiple agents were asking to do web searches. So now, now launching all four research agents in parallel. So this is off and I've got, you know, my technical architecture agent, I've got this other agent that these are both doing web searches right now. So one is looking at like prediction market order book matching engine architecture. So

• 19:59Speaker 01

  this one's learning about engine architecture for prediction markets. This one's looking at poly market, how it works, binary prediction market mechanics. And then I've got the UX design is doing some design research and then we've got some test research. Okay, now it's going to go to poly market and let's really hope the poly market doesn't block it. So I'll make things harder for it. Meanwhile, over here, this has discovered, Codex has figured out the repo is empty. So it's going to scaffold it from scratch. And it is starting to, I'm now wiring the core market math and trading engine. So it's interesting, right? So you've got Codex is out

• 20:38Speaker 01

  here building and is like building the engine. With Opus 4.6, it still has agents out there like doing research work.

• 20:48Greg Isenberg?

  Yeah, you really start to see just like how different they really are.

• 20:51Speaker 01

  Yes.

• 20:52Greg Isenberg?

  As they make progress.

• 20:55Speaker 01

  Yeah. Like I said, I haven't tested it before. So we don't know how long it'll take each of these.

• 21:00Greg Isenberg?

  Totally. Yeah. And I think like, I guess one question I have is like, is one, is one model better for being more of a beginner, but non-technical vibe coder? Or, you know, doesn't really matter.

• 21:16Speaker 01

  Yeah, it's a good question. I mean, I think the fair answer would be probably Codex because Codex edged out Opus 4.6 a little bit on some of those coding benchmarks and is kind of known for writing better production code. Probably Codex in that way. At the same time, one of the downsides, and like I said, this is, I could only do this in a totally balanced way because they're so different. You know, at the same time, for a vibe coder, knowing when to interject and stop Codex and say, oh, wait, you're doing this this way. Can you instead look at doing it this way? They're probably not going to know how to do that.

• 22:01Speaker 01

  Right? And so that's where maybe Opus 4.6 is better, where you could say, okay, spin up four or five agents and let them work with each other. Right?

• 22:11Greg Isenberg?

  Yeah.

• 22:13Speaker 01

  Okay, Codex is done. All right. So Codex built a competitor to Polymarket in 3 minutes and 47 seconds.

• 22:22Greg Isenberg?

  And to be clear, Polymarket's a multi-billion dollar company.

• 22:25Speaker 01

  Yeah, I don't think this will work quite as well. But, but we'll see, let's see. So let's just check out if it worked first. I'll let this keep running here. So, you know, it'll tell you at the end here, it actually did the testing. So you can see it built a test suite. So it has an LMSR math unit test suite, an engine behavior unit test suite, and an API integration test suite. And it passed with 10 out of 10 tests. As far as what it built, it has this core LMSR market maker engine. So coherent pricing, slippage, bound loss behavior, domain trending engine.

• 23:07Speaker 01

  It built a REST API router, which is kind of interesting because I didn't tell it that it would have to build, obviously, any of this in any way. It figured out the architecture on its own. Clean responsive front end. All right, well, let's see, let's see if it is actually. So let's go here. I'll let this keep running. This has got these four agents just running away here. And.

• 23:31Speaker 01

  I'm in here. I'm going to do 1pm test.

• 23:38Speaker 01

  All right, test 10 past 10. That looks good to me.

• 23:42Greg Isenberg?

  Oh, yeah.

• 23:42Speaker 01

  1pm start. All right, it's running. Uh, let's see. Okay, here we go. Uh, so

• 23:53Speaker 01

  this looks like it has the ability. So let's, Greg, let's make you, we'll make you the first trader.

• 23:58Greg Isenberg?

  All right.

• 24:01Speaker 01

  Say add. Okay. You got a thousand bucks.

• 24:05Greg Isenberg?

  Okay.

• 24:05Speaker 01

  There we go. Not bad. All right. What, uh, what market do you want to create?

• 24:11Greg Isenberg?

  Um, well Bitcoin, I think as we speak has crashed to what 63,000 or something?

• 24:17Speaker 01

  Something like that, yeah.

• 24:18Greg Isenberg?

  So I do like the, I mean, will BDC be about, you know, be above 110K by?

• 24:25Speaker 01

  Yeah. Okay.

• 24:28Greg Isenberg?

  By desk 31.

• 24:29Speaker 01

  That's pretty, pretty good. Yeah. Okay.

• 24:31Greg Isenberg?

  Pretty good.

• 24:31Speaker 01

  So let's.

• 24:32Greg Isenberg?

  I mean, that's almost double.

• 24:34Speaker 01

  Yeah, that'd be pretty good.

• 24:37Greg Isenberg?

  It depends. It depends when you bought it, you know, if you bought it at, uh, 125K, then.

• 24:43Speaker 01

  Yeah.

• 24:44Greg Isenberg?

  You're not so happy about it.

• 24:47Speaker 01

  Let's see. So then I don't know, I don't even know what resolution criterion source would be. I mean, I think I know what it's getting at, but I guess you could say like, uh, uh, why don't we say use coin market, uh, cap as the source and resolve by looking at the price uh, on the last day of December, uh, just before midnight, I guess.

• 25:16Greg Isenberg?

  Yeah, I guess like.

• 25:19Speaker 01

  Uh, the price of BTC. Yeah. All right. Okay. It looks like it's okay. So we've got it now. So we'll use coin market cap. Okay. So then you could do a yes 50%.

• 25:35Speaker 01

  So what do you think? Yes, yes or no?

• 25:39Greg Isenberg?

  I mean, this isn't financial advice. This is just purely

• 25:43Greg Isenberg?

  purposes. But I think I think so. I think that

• 25:47Speaker 01

  All right, that's a yes for Greg. Buy. Um, let's see. How many shares you want to buy? You've got 1,000 bucks.

• 25:56Greg Isenberg?

  I want to put it all. I'll put it all on this.

• 26:00Speaker 01

  I don't know how much it is per share. Let's see if it's 1,000, if that's right. Okay, yeah. Okay, trade executed. Okay, so I mean, it seems like it built something, you know, as a prototype relatively functional here. I guess that it actually has decremented. So, okay, 1,000 shares was not that ended up being you know, about $24 that you spent. So you've got more money if you wanted to create another market. But it worked. It's not returning in air. It shows the volume here. Interesting. All right. So let's go back. Let's see. So far, so good with that, I'd say.

• 26:38Speaker 01

  Let's see what's going on here. So we've got, okay. So first off, look at how many tokens. People have been talking about how token hungry Opus is, and it's very token hungry. Each one of these agents has used over 25,000 tokens.

• 26:55Speaker 01

  So let's see, though. So they finished, right? The the the technical research around architecture is done. Prediction market research is done. The UX design research is done. The testing strategy is done. Now it's going to go and build. So it's writing the package JSON.

• 27:17Greg Isenberg?

  Did you see the ad that Anthropic launched about ads?

• 27:24Speaker 01

  Yes, I watched them all. They're hilarious. Although, actually, I guess Sam was not very happy about them today. I saw a tweet from Sam that was less than happy. So I found them hilarious, but I also understand his side as well.

• 27:38Greg Isenberg?

  So basically, it seems like Anthropic is sort of anti ads for now and

• 27:45Speaker 01

  Yes.

• 27:45Greg Isenberg?

  Chat GPT is going to be introducing ads.

• 27:47Speaker 01

  Yes.

• 27:48Greg Isenberg?

  And, you know, when I'm, when I'm watching this and I'm seeing you're going through 25,000 tokens, 25,000 tokens, 25,000 tokens.

• 27:56Speaker 01

  Yeah.

• 27:56Greg Isenberg?

  I'm like, yeah, of course Anthropic doesn't really.

• 27:59Speaker 01

  Exactly.

• 28:01Greg Isenberg?

  You know?

• 28:02Speaker 01

  Yeah, yeah. I mean, this is literally, I mean, if you add that all up, you're talking about over 100,000 tokens.

• 28:08Greg Isenberg?

  Yeah.

• 28:08Speaker 01

  Used in doing this. So I think that's one of the very good things for like investors in Anthropic. Right? Is with agents and agents now being, I think probably the new killer feature in Opus.

• 28:24Greg Isenberg?

  Yeah.

• 28:24Speaker 01

  You're going to take whatever token usage and multiply it by the number of agents.

• 28:27Greg Isenberg?

  Exactly. It's actually really smart. And I wonder if that was like the thinking, like they're like, how can we get people

• 28:35Speaker 01

  Yeah.

• 28:35Greg Isenberg?

  More tokens? Oh, we'll just like spin up agents and we'll design it like that.

• 28:40Speaker 01

  Yeah.

• 28:40Greg Isenberg?

  Or did they think like, okay, how can we design a system that is best for the use case? And then they're like, okay, then we'll monetize it like this. I don't know.

• 28:48Speaker 01

  Yeah. Yeah. Probably a combination of the two. I can tell you, I've never used so many tokens in one day as today.

• 28:55Greg Isenberg?

  So.

• 28:56Speaker 01

  It's working.

• 28:57Greg Isenberg?

  A hundred, 100,000 tokens is like roughly how much in, in US dollars?

• 29:02Speaker 01

  I don't know, because I have, I have a Claude Max.

• 29:06Greg Isenberg?

  Yeah.

• 29:06Speaker 01

  Plan. So, yeah, so I'm not paying, we're not seeing it hit any limits right now, right? So I'm not paying more than $200. I can tell you that.

• 29:16Greg Isenberg?

  Yeah. My guess is it's, you know, we're talking like in, in the $200 Max plan, do you remember how many, how many tokens you get approximately?

• 29:29Speaker 01

  That's a good question. Um.

• 29:32Greg Isenberg?

  Let me check.

• 29:32Speaker 01

  Fire up, let me fire up Claude and ask it. Let's see here. How many tokens do I get? Estimate. Let's see. Okay. So here you go. Estimate, estimates. So 45 million tokens per month of SONET. But let's see. What is your estimate for Opus?

• 29:57Speaker 01

  Or six.

• 30:00Greg Isenberg?

  It's like, they don't really want you to know.

• 30:02Speaker 01

  No, they're trying to make it a little harder. Okay. But yeah, they're not even going to tell me actually, they're just gonna say, there's no public data. It's very new. Opus is roughly 5X more expensive. So then, uh, if it's 45 million, that's 5X or 10 million is probably the answer about, right?

• 30:19Greg Isenberg?

  Yeah.

• 30:19Speaker 01

  Yeah.

• 30:21Greg Isenberg?

  So then if we're doing, you know, quick math, let's just say we spent 100,000 tokens.

• 30:28Greg Isenberg?

  You know, 100,000 divided by 5 million is, you know.

• 30:34Speaker 01

  Well, we're gonna spend more than that because look at this. We're now over 17,000 tokens on top of that in this next build.

• 30:42Greg Isenberg?

  Okay.

• 30:43Speaker 01

  So, but still, let's say, you know, even if we use a million tokens, building a competitive supply market right now, we're still only using a 10th of what a buddy can do. That's not terrible.

• 30:55Greg Isenberg?

  No, I mean, it's $20, which is like the price of a cocktail in Miami.

• 31:00Speaker 01

  Yeah, yeah, exactly. Yeah, yeah. So let's see, but now, as I say, I'm watching the tokens creep out.

• 31:10Speaker 01

  Alright, so it's building the API routes now.

• 31:15Greg Isenberg?

  I have a feeling this is going to be a better end result.

• 31:18Speaker 01

  I was actually just gonna say that. This feels, and maybe it's just because there were four agents that were doing all the work beforehand. And now it's doing the work. It feels like we're going to see something very different when we, uh, when we load what it builds.

• 31:35Greg Isenberg?

  Yeah. I don't think we gave it any like design, like visual design, any, you know, so.

• 31:41Speaker 01

  Nope.

• 31:42Greg Isenberg?

  Do you recommend for folks to just like sort of get the MVP out, out, play around with it on local host? Yeah. Click some buttons and then sort of update with the visual design from there?

• 31:55Speaker 01

  It's a good question. I do like 50/50. Sometimes if I have something in mind, especially if I want something like on brand with something like suppose I'm building something that is going to be in the like OpenClaw Multibook ecosystem, I would probably say, "Hey, I want to design a site that, you know, looks somewhat similar to or is inspired by, you know, openclaw.ai and multibook.com, right? Take a look at those sites and get inspiration. These models are great at doing stuff like that."

• 32:33Greg Isenberg?

  Cool.

• 32:33Speaker 01

  I'm really excited to see what this is doing though. I think we're now like well over 200,000 tokens. They show up, I can tell. But we're not at 10 million.

• 32:44Greg Isenberg?

  We're not hitting any limits. You know, we don't have to take out a second mortgage on our...

• 32:49Speaker 01

  Yeah, here we go. Yet. Yeah, it's still going though.

• 32:55Speaker 01

  I guess, you know, we can tell, like, here's an interesting thing in a comparison. Like, this is still going. Why don't we say like, "The design..." Because the design looked kind of bland to me, right?

• 33:06Greg Isenberg?

  Yes, it did.

• 33:12Speaker 01

  Can you spruce it up and make it look nicer? Because like, we may as well have Codex working away too, right?

• 33:19Greg Isenberg?

  Yeah. So you didn't really give it any like specific, "It should look like square.com."

• 33:25Speaker 01

  No, no.

• 33:26Greg Isenberg?

  So we'll see. We'll basically see if Codex, if, you know, if 5.3 has a little bit of taste.

• 33:35Speaker 01

  Yeah. Yeah. So that's what it's saying now. It's saying, "Okay, I'll upgrade the visual system without changing functionality. Stronger typography, richer color direction, better card hierarchy, and purposeful motion." I don't know what that means, but we'll find out. All right. Okay. So now it's, now it's editing index.html. It looks like it's going to add motion hover polish. Okay. Totally. Look at that. "This current task for over 30,000 tokens, building the front-end UI." Okay, it's done. Yeah, Codex is fast, by the way, right? I mean, that's pretty darn fast.

• 34:11Greg Isenberg?

  Yeah.

• 34:12Speaker 01

  So we should be able to just go here. It should have already automatically reloaded, but okay. All right. I mean,

• 34:20Greg Isenberg?

  I mean,

• 34:21Speaker 01

  not that different.

• 34:22Greg Isenberg?

  Not that different. It's, I think, um, can I try something?

• 34:28Speaker 01

  Yeah, go for it.

• 34:29Greg Isenberg?

  I'm gonna say, I would say, okay, thank you. But this was a minor design refresh. I'm looking, I'm looking for a major one.

• 34:41Speaker 01

  There you go. Yeah.

• 34:42Greg Isenberg?

  And then I'm gonna say, pretend

• 34:48Greg Isenberg?

  you are Jack Dorsey.

• 34:51Speaker 01

  There you go.

• 34:52Greg Isenberg?

  And how would he design this website to be clean, elegant, and full, and full of interesting interactions?

• 35:04Speaker 01

  Yeah, great. Yeah.

• 35:07Greg Isenberg?

  Jack Dorsey, for people that don't know, co-founder of

• 35:11Greg Isenberg?

  formerly known as Twitter and Square Block now. He's just got, he's got, he's a design guy. I don't know. He's first one guy that came to mind or first person that came to mind.

• 35:23Speaker 01

  Yeah, that's a good one. That's a good prompt. Let's see.

• 35:27Speaker 01

  So I'll do a full visual re-architect, not an incremental tweak, new layout language, stronger typography, monochrome first palette, interesting, and their action-driven cars, duh, duh, duh. Okay. You know what's interesting is it didn't, I would have kind of hoped, and maybe, you know, we're not quite at AGI yet, I would kind of hope that it would say, let me go find some art. Like, if you told me that, Greg, like, hey, Morgan, can you read it? I would be like, yeah, let me go look at some articles about Jack Dorsey's design aesthetic.

• 36:00Greg Isenberg?

  Exactly.

• 36:02Speaker 01

  I'm surprised it's not doing that. Instead, it's going like, I am assuming it knows who Jack Dorsey is, although I don't know if it actually does. It just seems like it, it's, it's really just taking like this part of your question and going, oh, okay, major refresh, I'll do that.

• 36:18Greg Isenberg?

  Well, can't, can't you ask it? Can't you say, do you know who Jack Dorsey is?

• 36:22Speaker 01

  Let's see, I can actually, I'm supposed to be able to, in the middle, cut it off. So let me see. Yeah. Do you know who Jack Dorsey is? Let's see. Okay, so here we go. This is the midstream test.

• 36:39Speaker 01

  Thinking about it. 43,000 tokens over here. Okay.

• 36:45Greg Isenberg?

  There you go.

• 36:46Speaker 01

  Okay, here we go. Yes, Jack Dorsey is the co-founder of Twitter, former of Dog and Square. Okay, with a design style that's typically minimal restraint and interaction focused.

• 36:56Greg Isenberg?

  Beautiful.

• 36:57Speaker 01

  Okay. All right. So touche, it showed us.

• 37:01Greg Isenberg?

  Yeah.

• 37:03Speaker 01

  Now, here's the weird thing. It looks like it's.

• 37:05Greg Isenberg?

  Is it complete or do we have to say?

• 37:07Speaker 01

  Complete, like, right when I was saying that, like, are you done or did you stop because I asked a question?

• 37:14Greg Isenberg?

  Yeah, this is really interesting.

• 37:18Greg Isenberg?

  I will say, like, that, oh, I paused when you asked the question. The major redesign. Mostly.

• 37:25Speaker 01

  If you want to resume, oh, so that's weird. So you asked a question, it just stops, but like, so like, yes, of course, continue.

• 37:33Greg Isenberg?

  Yeah.

• 37:34Speaker 01

  Okay.

• 37:35Greg Isenberg?

  So that's, that's actually some weird UX. Like, it obviously should just continue after.

• 37:39Speaker 01

  Right? Yeah. I would assume that. It's such a weird thing, because it said, yeah, if you want, I'll resume now.

• 37:47Greg Isenberg?

  I will say, I, I do like that you can, in midstream, like, kind of edit things.

• 37:53Speaker 01

  Yeah.

• 37:53Greg Isenberg?

  Like, that's how my brain works.

• 37:55Speaker 01

  Totally. Yeah, yeah.

• 37:59Speaker 01

  True, this is using a ton of tokens. It is amazing to see the detail. I mean, this, this should be a work of art, whatever cycle this comes off.

• 38:11Speaker 01

  All right, this is done. So now, let's see, we can go back to this. And, uh, okay. I mean, I'm not blown away, but it's okay.

• 38:22Greg Isenberg?

  Opinions become price in milliseconds. Trade conviction, not noise. Signal market is dying for fast thesis inter- iteration with transparent pricing. Okay. I mean, I, I would push it more, I think.

• 38:38Speaker 01

  Yeah. Yeah, I guess, you know, like.

• 38:41Greg Isenberg?

  I would say.

• 38:42Speaker 01

  Go for it.

• 38:43Greg Isenberg?

  I would say that's not the Jack Dorsey I know.

• 38:49Speaker 01

  I don't know Jack Dorsey.

• 38:53Greg Isenberg?

  I was looking for a Caps Lock major upgrade. That that might mean way more copy, way more images.

• 39:08Speaker 01

  Yeah.

• 39:10Greg Isenberg?

  Way more storytelling.

• 39:12Speaker 01

  Yeah, exactly.

• 39:14Greg Isenberg?

  Et cetera, et cetera.

• 39:15Speaker 01

  Yeah, I'll just say, seriously, take your time, go nuts.

• 39:19Greg Isenberg?

  Yeah.

• 39:20Speaker 01

  What are credits?

• 39:23Greg Isenberg?

  Famous last words, Morgan.

• 39:24Speaker 01

  I know, right? It's like, oh, perfect. Okay. That's like a signal within the OpenAI headquarters, like, we finally got someone.

• 39:30Greg Isenberg?

  Totally, it's a whale.

• 39:32Speaker 01

  All right. So Opus has finished.

• 39:35Greg Isenberg?

  Yep.

• 39:35Speaker 01

  I have no idea how many credits I would use, but probably, actually, let me ask it. How many, how many tokens in total did you use to put all of this together, including the four agents? And then we can, oh, it's using token stance.

• 39:52Greg Isenberg?

  That's so funny.

• 39:54Speaker 01

  Okay.

• 39:56Speaker 01

  Doesn't know. Let's see. Okay, here we go. Okay, it's estimating. It actually doesn't know, which is weird, because it should know. Although, I wonder if I can actually do slash cost. Oh, here we go. Yeah, okay. Oh, it doesn't. Okay, no, no need to monitor cost. Okay.

• 40:15Greg Isenberg?

  Yeah, no need.

• 40:15Speaker 01

  They really don't want you to know.

• 40:19Speaker 01

  Okay, it's guessing 150, 250,000 tokens total.

• 40:22Greg Isenberg?

  Yeah, that's probably right.

• 40:24Speaker 01

  Okay, sure.

• 40:24Greg Isenberg?

  Yeah.

• 40:25Speaker 01

  Okay, so here's what it's done. So first off, one really interesting thing here is, you know, Codex created 10 tests, right? Opus created 96 tests. So definitely a lot, a lot more detail on the testing side. And it's called it Forecast, whereas Codex called it Signal Market. So different names. A polymark competitor is built and verified. Here's what each team member delivered. So the architecture technical lead decided modular monolith, NextJS14 app router, central limit order book.

• 41:05Speaker 01

  Database schema Russell API. Okay. The prediction market domain expert binary yes, no market where yes, no is always a dollar. Okay. Seated markets across crypto politics. Okay. The UX design lead dark mode trading platform pages. It is a green for yes, red for no. Okay. Testing QA lead did order book tests. Okay. So here's how the tests are breaking out. Order book test matching engine. Okay.

• 41:37Speaker 01

  All right. NPM run dub to start the app. So let's go in here. Oh, interesting. Okay. I don't want to say anything. Actually, I already given to it. What's your initial take?

• 41:49Greg Isenberg?

  I mean, my hello, Jack Dorsey, you know what I'm saying? This is, this is what I expected it to look like when we pushed codex.

• 42:00Speaker 01

  Yeah, me too.

• 42:02Greg Isenberg?

  This looks really clean. What happens when you hover over?

• 42:08Speaker 01

  Oh yeah. Look at that. Yeah. It's got hover states.

• 42:11Greg Isenberg?

  Hover state.

• 42:13Speaker 01

  It's actually got it organized like where those sports, you know, will the next tube will have over 120 million viewers? Will AI pass the Turing test by 2027? Will a movie go to three? It's got stuff in there. Yeah.

• 42:26Greg Isenberg?

  Yeah. This doesn't even, it doesn't feel like an MVP.

• 42:30Speaker 01

  Yeah. This is pretty wild, actually. And it created some stuff, you know, that we never talked to it about, right? Like a looter board, which it's already populated with some initial stuff. Portfolio section. Yeah. Interesting. So let's see now. So I mean, I'm, I'm more impressed with the, it was maybe worth the 150,000 to 200,000 tokens. Feeling better about it. Will SpaceX land humans on Mars before 2030? Only 8% of things, huh? Oh yeah. Look at this actually. Whoa.

• 43:05Greg Isenberg?

  This is insane, bro.

• 43:06Speaker 01

  That's clean. That's yeah. I wasn't expecting to click in and actually get a well-designed page like this. Huh. So if I were to do that, I had to sign in the trade. I don't know if I'm going to be able to sign in because I haven't set anything up. Let me just check.

• 43:21Greg Isenberg?

  Well, you can sign up. It says, don't have an account.

• 43:25Speaker 01

  Oh yeah, sign up. I don't know if it gets it all connected though. Let's see though. All right, I'm snagging. You know what, actually, I'm gonna take the username Greg.

• 43:33Greg Isenberg?

  Wow.

• 43:33Speaker 01

  Still username. All right.

• 43:36Speaker 01

  Let's see. Okay, so yeah, it's probably because I was gonna say the database isn't wired up yet.

• 43:43Greg Isenberg?

  Right.

• 43:43Speaker 01

  So I'm not, I'm not surprised that I would actually have to do. I wasn't expecting to do that. So.

• 43:49Speaker 01

  But I get it. I mean, it's clean. This is pretty neat.

• 43:52Greg Isenberg?

  Yep.

• 43:53Speaker 01

  Yeah. All right, let's see. So then can this, all right, we've given, I don't know about you, but this is the last chance I'm gonna give Codex.

• 44:00Greg Isenberg?

  Yeah.

• 44:00Speaker 01

  On the design side. It's out of, it's out of opportunities here. So.

• 44:06Speaker 01

  Oh, here, it's funny though, in the end, it kind of, it's, it's acting a little bit like data from Star Trek.

• 44:11Greg Isenberg?

  Yeah.

• 44:12Speaker 01

  When you question, what are credits? In this context, credits usually means.

• 44:18Speaker 01

  That's good though. Okay, so let's see. Let's take a look and see. All right, here we go. The new version of Signal Market?

• 44:26Greg Isenberg?

  Boom.

• 44:27Speaker 01

  Oh. Oh, okay.

• 44:31Speaker 01

  This is getting a little bit interesting. Let's see here. Read the manifesto.

• 44:36Greg Isenberg?

  Yeah, I mean, I don't hate it.

• 44:39Speaker 01

  It's definitely better. Yeah.

• 44:42Greg Isenberg?

  I mean, it's. It's got a lot going on.

• 44:46Speaker 01

  Yeah.

• 44:47Greg Isenberg?

  But it's different. Terminal. It's different than a.

• 44:55Greg Isenberg?

  Yeah, different than any sort of prediction market app I've seen.

• 45:00Speaker 01

  Yeah.

• 45:00Greg Isenberg?

  From a UX perspective. I just feel like this is just so clean, though.

• 45:03Speaker 01

  This is so good. It's so fast.

• 45:05Greg Isenberg?

  Yeah.

• 45:06Speaker 01

  Yeah, I mean, I, I would say, you know, like I said, I'm not gonna say which one is, it's not that Opus is better than Codex or vice versa, but I would say in this test, Opus one.

• 45:16Greg Isenberg?

  Yeah, in this test, Opus one.

• 45:18Speaker 01

  Yeah.

• 45:19Greg Isenberg?

  That's just the truth.

• 45:21Speaker 01

  Yeah. Yeah, but we could give another, I mean, you know, you never know. Like, I think that what's interesting about this is, I mean, Codex built it like, I don't know how much faster. We can look at the timing on this video, but like 20 times faster or something, right?

• 45:37Greg Isenberg?

  Yeah, yeah.

• 45:38Speaker 01

  Yeah.

• 45:39Greg Isenberg?

  Well, anything else you wanted to cover? I don't think we'll have time to do another example, but anything else you want to cover between, you know, that you want to leave people with?

• 45:49Speaker 01

  Yeah.

• 45:52Speaker 01

  Let me see if there is anything else in here. I covered the adaptive thinking. Oh, I guess just on the orchestration, I would say, you know, this is a feature I'm probably the most excited about with Opus. And clearly, we saw in this example, it working really well. Just make sure to look at the documentation. It's all in the docs now. And it gives some examples as well. Because it has this idea of like compare with subagents of like context and communication and coordination and kind of breaks this down. And then it has like a sample prompt, more of the details on the display mode. There's a lot of other stuff that I didn't go into there. That's probably what I would leave people with. Because I think a lot of people are going to want to dive in and use

• 46:37Speaker 01

  use agents with Opus 4.6 and they've got pretty good details on all little tweaks that you can make with it.

• 46:45Greg Isenberg?

  Amazing. Well, Morgan, I can't thank you enough for coming on. I hope people love this episode. I love talking to you. You get

• 46:54Speaker 01

  Thank you for having me, Greg. It's a total honor.

• 46:57Greg Isenberg?

  You're, uh, yeah, it's just, I love how clearly you communicate and, and, and to technical people, but also non-technical people and you're criminally under-followed. So I'm going to include

• 47:10Speaker 01

  I appreciate that.

• 47:11Greg Isenberg?

  Links where you can find Morgan and follow him on X. He talks a lot about vibe coding over there. And Morgan, any, anything else you want to, you know, places that you want to leave people to go and check you out?

• 47:25Speaker 01

  Yeah, I mean, I'm the co-founder and CTO of Bold Metrics. I'll just give a little plug for us. We have a AI technology that's used by apparel brands and retailers. So if you're shopping online and want to find the right size, you see a Find My Size button. We a lot of times power that and have really powerful machine learning models that update and adapt over time to help people find the right size and give Lots of really interesting data, lots of amazing brands and retailers that you probably all know and love. And then me and my team, you know, we're using all of this tooling. Like I had a meeting with my team this morning about Opus 46 and Codex.

• 48:06Speaker 01

  And I've given everybody access to both of these. And I actually have multiple teams of mine that are trying current things we're working on and are actually testing with each to see which performs better. So, you know, the one thing I encourage all engineering teams to do is like, and engineering leaders to do is like, let your teams loose with this stuff. Let them try it. You know, some of this stuff is really cutting edge and really performant and gives us the opportunity to do better, more creative work.

• 48:34Greg Isenberg?

  Yeah. Stop listening to us right now. Like

• 48:37Speaker 01

  Yeah, exactly.

• 48:40Greg Isenberg?

  X out of this YouTube or Spotify link, but actually give us a like, a comment and subscribe. Let us know if you like this episode. Morgan, thanks again for coming on the show. This was a lot of fun. See you.

• 48:52Speaker 01

  Craig, thank you so much. Total honor.