Modern Creator
Peter Yang · YouTube

How This Non-Technical Founder Mastered Agentic Engineering

A 47-minute conversation with Matt Van Horn, who ships viral open source projects without reading code or even the plans his agents write.

Posted
5 days ago
Duration
Format
Interview
educational
Views
15.6K
490 likes
Big Idea

The argument in one line.

The bottleneck to shipping is not coding skill — it is the discipline to give an agent a structured plan first, trust it to execute without babysitting, and stress-test the output with a single adversarial question instead of reading every line.

Who This Is For

Read if. Skip if.

READ IF YOU ARE…
  • You build products or ship features but are not an engineer, and want to see how non-technical founders run real agentic workflows today.
  • You use Claude Code, Codex, or Hermes and want your agent loops faster and less prone to lazy output.
  • You want to contribute to CLI tools or open source projects without deep CS knowledge.
  • You want a concrete example of compound engineering (ce plan / ce work) in practice, not in theory.
SKIP IF…
  • You are looking for a step-by-step tutorial — this is a conversation, not a course.
  • You already have a mature agentic engineering system and want frontier technical depth.
TL;DR

The full version, fast.

Matt Van Horn ships production-quality software without reading code or even the plans his agents write. His core loop — /ce plan then /ce work via Compound Engineering — forces the agent to produce a structured plan before executing, which defeats agents' natural tendency toward lazy shortcuts. He runs 6 agent windows simultaneously and asks one adversarial question per session to verify scope. His most viral project, the Printing Press, auto-generates a CLI + agent skill + MCP for any website by HAR-sniffing hidden APIs and absorbing community GitHub projects — no official API required. He has merged PRs into Python, Vercel, OpenCV, and 200+ other repos, mostly via automated overnight systems, and credits the resulting network as the most valuable asset he has built.

Free for members

Chat with this breakdown — free.

Sign in and you get 23 free chat messages on us — ask for the hook, quote a framework, find the exact transcript moment, generate a markdown action plan. Bring your own key when you want unlimited.

Create a free account →
Voices

Who's talking.

00:30hostPeter Yang
00:30guestMatt Van Horn
Chapters

Where the time goes.

00:0001:08

01 · Cold open

Matt states he does not read code; Peter frames the interview premise.

01:0806:55

02 · The CE Plan loop

Compound Engineering's ce plan / ce work cycle; the real morning Agent Cookie example; why the plan is for the agent, not you.

06:5510:55

03 · Non-engineer origin story

Matt's background as a self-described nerd with no CS degree; arriving in SF in 2007; BCAC inflection point Thanksgiving 2024.

10:5519:00

04 · Printing Press origin

Failed 'build a programming language' idea -> Printing Press; Peter Steinberger's SQLite-backed CLI pattern; power-user persona design.

19:0024:54

05 · Printing Press demo

Live Papa John's CLI print; CLI vs MCP auth reliability; what gets printed (CLI + agent skill + MCP for every tool).

24:5432:46

06 · Open source addiction

How Matt started contributing; the adrenaline of a merged PR from Peter Steinberger; automated overnight contribution systems.

32:4640:00

07 · GitHub contributions deep dive

last30days 20K stars, Printing Press 3K stars, Python CPython cross-language method suggestion feature; banned from Rust.

40:0047:45

08 · Fear of shipping

Two finished projects Matt fears launching; the founder simulation game; the build-for-yourself floor.

Atomic Insights

Lines worth screenshotting.

  • Agents are lazy by design — they want to make you happy as fast as possible with as few tokens as possible. The only fix is forcing them to write a plan first.
  • The plan is not for you to read. It is for the agent to stop cutting corners.
  • Ask one adversarial question to verify scope after the plan is done — then let it run without reading anything else.
  • You do not need a CS degree to merge PRs into Python, Vercel, or OpenCV — you need agent discipline and good adversarial prompting.
  • CLIs beat MCPs for auth reliability because CLIs encode exactly how to refresh a token; MCPs externalize the auth burden to the human.
  • The Printing Press discovers every API via HAR sniff — Kayak, Google Flights, MidJourney, Suno — none of which have official APIs.
  • Open source contribution is the fastest recruiting tool right now: contributors get access to the secret network of the best agentic engineers.
  • HAR sniffing finds every secret API a website calls through the browser — those endpoints often go unchanged for a decade (see: Dominos).
  • CLI + agent skill beats MCP alone because the skill tells the agent when and how to call the CLI without relying on luck.
  • Peter Steinberger's pattern of building SQLite-backed CLIs beats Google's official CLI in head-to-head agent benchmarks because of local-first data and power-user persona design.
  • Build for yourself first — a tool that solves your own problem has at least one committed user and forces genuine engineering instead of trend-chasing.
  • The fear of public rejection is real even after repeated viral launches — the floor of 'at least I solved my own problem' is how experienced builders manage it.
  • The cost of testing an idea has dropped to near zero — bias toward running more experiments, not waiting for certainty.
  • Most of Matt's open source contributions happen while he sleeps — fully automated overnight systems built over months of iteration.
Takeaway

You do not read the plan -- the plan is for the agent.

WHAT TO LEARN

The entire shift in agentic engineering is realizing the structured plan is not your checklist to review -- it is the agent's instruction to stop cutting corners.

02The CE Plan loop
  • Force every agent session to start with a planning step before any execution; agents without a plan default to lazy shortcuts that look complete but are not.
  • Ask one scoped adversarial question after the plan is written -- 'are you only touching the right codebase?' -- then let it run without reading the rest.
04Printing Press origin
  • Build CLIs that download data into a local SQLite database rather than making live API calls on every query -- this is the primary advantage that makes community CLIs beat official ones.
  • HAR sniffing in the browser reveals every API a website calls, including undocumented secret endpoints that often stay stable for years.
05Printing Press demo
  • CLI tools beat MCP integrations for auth reliability because CLIs encode re-auth logic internally; MCPs externalize the auth burden to the human.
  • Run multiple agent windows simultaneously on different ideas; most experiments are cheap enough that running a bad one is negligible cost.
06Open source addiction
  • Open source contribution is the fastest path to entering the network of top builders -- a merged PR creates a relationship that a cold message never does.
07GitHub contributions deep dive
  • When evaluating whether an agent's work is correct, do not read the code -- ask the agent to play devil's advocate and argue against its own approach.
08Fear of shipping
  • Build for yourself first; a tool that solves your own problem has at least one committed user and forces genuine engineering instead of trend-chasing.
  • The fear of public rejection is real even after repeated viral launches -- the floor of 'at least I solved my own problem' is how experienced builders manage it.
Glossary

Terms worth knowing.

Compound Engineering
An agentic coding tool that provides /ce plan and /ce work commands, forcing the agent to produce a structured execution plan before beginning any implementation work.
CE plan / CE work
The two-step Compound Engineering loop: /ce plan has the agent research and write a plan.md; /ce work executes that plan. Running plan before work prevents agents from taking lazy shortcuts.
HAR sniff (HAR sniffing)
Recording the browser's HTTP Archive (HAR) while using a website to capture all network requests, including undocumented internal APIs the site calls. These secret endpoints can then be used programmatically.
Printing Press
An open source tool (printingpress.dev) that generates a production-ready CLI, agent skill, and MCP for any website or API from a single prompt — using HAR sniffing, official API discovery, and GitHub community project ingestion.
Agent Cookie
A tool Matt built that keeps browser cookies in sync across multiple machines over Tailscale, so agents running on different computers stay authenticated to all services without manual re-login.
BCAC
Matt's shorthand for 'Before Claude and Codex / After Claude and Codex' — the Thanksgiving 2024 inflection point when these tools stopped being toys and became capable of shipping real products.
last30days
Matt's open source GitHub project that became the #1 trending repository on GitHub with 40K+ stars — a tool that analyzes the last 30 days of activity across a codebase or community.
OpenClaw
An agentic computing platform built by Peter Steinberger, known for a disciplined pattern of building SQLite-backed CLI tools for every service rather than relying on MCPs or official APIs.
Resources

Things they pointed at.

01:27toolCompound Engineering
03:11productAgent Cookie
11:11productPrinting Press
13:00toolOpenClaw (Peter Steinberger)
14:51productlast30days
08:58productRiverside
17:16toolGoogle Flights / Kayak secret APIs
32:40toolVercel Agent Browser
38:13toolPython CPython (cross-language method suggestion PR)
Quotables

Lines you could clip.

02:17
Agents have a propensity to be lazy. They wanna make you as happy as possible as fast as possible with as few tokens spent as possible.
Tight, counterintuitive framing of a universal LLM problemTikTok hook↗ Tweet quote
15:39
What if anyone could print their own CLI about anything?
Founding question that summarizes an entire product categoryIG reel cold open↗ Tweet quote
17:50
Every API has a secret identity.
Punchy line from the Printing Press tagline, true and surprisingnewsletter pull-quote↗ Tweet quote
46:58
Even if I had no users of agent cookie, like, if no one used it, I get value out of it. Like, it helps my agents be so much better.
Clean articulation of the build-for-yourself philosophyIG reel cold open↗ Tweet quote
43:13
I'd say I'm the most incompetent person that I've met in terms of actual skills to success ratio.
Disarmingly honest self-assessment from someone with 20K GitHub starsTikTok hook↗ Tweet quote
Topic Map

Where the conversation goes.

01:0806:55denseCE plan / CE work loop mechanics
06:5510:55steadyNon-technical founder background
10:5519:00densePrinting Press concept and origin
15:3819:00denseHAR sniffing and secret APIs
19:0024:54steadyLive demo and CLI vs MCP
24:5432:46denseOpen source community and networking
32:4640:00denseGitHub contribution stats and Python feature
40:0047:45steadyFear of shipping and founder psychology
The Script

Word for word.

Read-along

Don't just watch it. Burn it in.

See every word as it's spoken — crank it to 2× and still catch all of it. The same dual-channel trick behind Amazon's Kindle + Audible.

00:00I certainly don't read any code. I don't even know how to read any code. I thought you were actually an engineer.
00:05I'm not an engineer. I don't know what you call me. Biggest hack, like, if you do one thing is my favorite tool for building anything is compound engineering.
00:14Somehow, I'm able to ship things of value, which is crazy and weird and still blowing my mind. What if anyone could print their own CLI about anything? They found all the secret APIs by sniffing, and it worked.
00:30Hey, everyone. My guest today is Matt, who is arguably one of the best and most prolific AI builders that I know. And today, he's gonna walk us through his complete agentic AI well, I guess, agentic engineering system that went viral on X.
00:43And I think, uh, you know, from what I've seen from Matt, I think this interview is gonna blow your mind. So welcome, Matt. Thank you.
00:48Excited to be here. Let's let's talk about your like, you posted this article, right, that had, like, I think, 20 tips or something about building with AI agents. Do you wanna share an overview of it?
00:56Sure. Yeah. I'll I'll share the screen you made.
00:59Alright. I didn't make the slide you did. This is Matt's Anti Genetic System.
01:02I think you should present this, Peter. No. No.
01:05No. I I I I made this based on your article. So, I mean, there's a lot of tips on here.
01:10Lot A of tips on here. But, like, what what are some high level themes? You know?
01:13Like, maybe one high level theme, like like you told me, is, like, you don't actually read you don't actually read the plans. What are some high level themes? Yeah.
01:20Sure. So I think the the biggest
01:22hack, like, if you do one one thing is the my my favorite tool for building anything is is compound engineering.
01:31And I'm it does compound engineering does a lot of things. Every has done an incredible job.
01:37Trevin's done an incredible job, uh, Karen. And but the the killer feature is CE plan and CE work.
01:46And what that does is if I ever have an idea like, I'll okay. Can give an example of an idea I had this morning.
01:53But, uh, if I ever have an idea, I always just start with slash CE plan. Here's my idea. Uh, I'm usually using voice.
02:02And what it's gonna do is it's gonna write a plan dot m d file, and that that's coming up with the execution plans for the agent. And the reason this matters versus just talking straight to Claude or to Codex is agents have a propensity to be lazy.
02:20Like, they wanna make you as happy as possible as fast as possible with as few tokens spent as possible. And by writing down a step by step systematic plan, it sets itself up so it's not lazy.
02:35And so for anything, I always say, CE plan, CE work. CE plan, CE work. And back in the day, and by back in the day, I mean, like, two months ago, I used to always read the plans.
02:47Right? My last I've only written two of these agentic engineering articles, and the the last one said, oh, we'll open the plan in your editor.
02:55Don't look at the code, obviously, but you should still look at the plan. And in the last two months, I found there's no reason to look at the plan and that I can just ask questions to to the agent.
03:07So I'll give an example of something I worked on this morning. So I I I launched a product, I think, this week.
03:12Yeah. I guess this week on Monday called Agent Cookie, which allows your so if you have a main Mac laptop that you're using and a Mac mini that you use for either Hermes or OpenClaw, the cookies are out of sync, and so you're logged out all the time.
03:32And so I built something over Tailscale that makes your cookies perfectly in sync. And so this morning, I was using another product of mine that I built with Trevin called the printing press, and it often it's creating CLIs out of anything.
03:47And so I'm trying to create a TaskRabbit CLI right now. Can get into that in a second why that's interesting. It's a it's a human goat CLI, but that's a side thing.
03:56And often on my own machine, my agent is logged out.
04:01And so I said this morning, CE plan, take a look at the printing press, and take a look at agent cookie to unrelated projects of mine Yeah.
04:12And figure out what printing press uses for harsh sniffing when creating a new CLI or agent skill and how agent cookie currently works and figure out how all my agents locally so that's for sales agent browser.
04:31That's browser use. That's, uh, I think, playwrights. How they are always logged in to everything, and then it's like, okay.
04:39I made a plan. And then I didn't load the plan because you should never look at the plan because you gotta move fast. You gotta move fast.
04:45But I had one question. I had one question for my agent this morning. I said, you're just modifying agent cookie.
04:52Right? You don't have to, like, make a change to Versal's agent browser or to playwright or to browser use.
04:59And he's like, nope. Just agent cookie. So imagine we work just in agent cookie.
05:02Like, great. Alright. CE work.
05:04Let's go build. And I I think it built it, by the way. I think it finished a few minutes ago.
05:08Haven't looked at it yet. Can you actually, like, go to your terminal and, like, show me what the plan looks like? I said you're not supposed to look at the plan, but sure.
05:15Let's let's see it. Yeah. I'm gonna try to send this between c plan and, uh, just, like, regular plan mode.
05:22Like, what's so great about c plan? Let's see what what happened here. So I said, c plan, can you dig into the printing press and how the agents use agent browser, browsers for hard sniffing, then dig into the agent cookie works and how to sync logged ins.
05:36Are you sure you're logged in? Office says I'm not. How can we fix that with agent cookie?
05:39And then I said, no. Sorry. I didn't mean that.
05:41Is it only agent cookie? And it's like, yes. I'm only modifying h kind of.
05:44Was like, alright. Great. Go.
05:45So that was, like, my one question without ever looking at this plan. Obviously, I could have read this plan Yeah. And maybe that would have created some value, but not really because I just had one question.
05:56If I was submitting a bunch of PRs to AgentBrowser
05:59and Playwright and modifying their code bases versus just making a PR to agent cookie. Yeah. So I guess to summarize, like, the difference between and I interviewed Karen too.
06:09The difference between CE plan, like asking the agent to plan is like, it does a ton of re research on your code base and online, and it also, like, writes a pretty comprehensive plan. Right?
06:19Like, it goes into a lot more detail.
06:21You know? So I don't even know because I don't read them is is the honest answer. But Okay.
06:25But the results are good, and and my products often work, they don't break. Yeah. Right?
06:30So, like, for me, it's just about the iteration cycle. It's like, okay. I asked you to make a thing, and you made a thing.
06:37And then it's like, okay. Does the thing work? Is it any good?
06:41And if it does, great. Thank you. Thank you, Kieran.
06:44Thank you, Combat Engineering. But I think the the big thing is I I don't have to worry about it because I've used it so much that I'm able to just trust it, and I know what types of questions lead to success and failure.
06:56Got it. So so you do this for, like, pretty much every new feature you wanna build, and then you see build. And then do you read the code, like, PRs and stuff, or do you kind of, uh, is there, like, a command for that too?
07:05Like, like, skill? So I certainly don't read any code. I I don't even know how to read any code.
07:11I I haven't written any type of code since high school, and that was probably HTML. I don't even think CSS was was that big when I was in high school. And maybe some AppleScript language type stuff.
07:23But that that that that's my engineering skill set. So I certainly don't do that. Uh, and then if I PR, if I'm submitting an external PR, then I I like to glance at it, obviously, and and hit submit and make sure I'm saying things that are normal.
07:40That that that's actually, uh, okay. Maybe I should have done more research. Okay.
07:43I I I thought you were actually an engineer, but it sounds like you actually
07:47you're a a GenTech engineer. Not not not a real engineer. I'm not an engineer.
07:50I don't know what you call me. I'm I I've called myself a nerd my whole life. Like, I remember when I got my first job at at DIG in 2007, and I I've been a nerd my whole life.
08:01Like, everyone would call me to, like, fix their Wi Fi. Like, I used to camp out for Macworld to see Steve Jobs speak when I was a kid. And and then I arrived in San Francisco in 2007, and I was like, I'm one of you.
08:13You're my people. Yeah. I'm so excited.
08:15And then they're like, do you have a CS degree? Are you an engineer?
08:19I'm like, no. I'm just a nerd like you. They're like, not so much.
08:23And so I I've always felt this, um, I don't know what the word is. Like, uh, like, there was always a negative signal.
08:32It's like, oh, Matt doesn't know. He doesn't understand.
08:35And, again, this this I still don't try and pretend like I know. I literally say, I don't read the plan. I don't know what's happening.
08:43And, um, and and so I obviously respect engineering and real real engineers, and I am not under any circumstance that. But somehow, I'm able to ship things of value, which is crazy and weird and still blowing my mind.
08:58This episode is brought to you by Riverside. I've used Riverside for years to record my podcast because it records in four k resolution each person locally, so the audio and video still comes through clean even if a guest's WiFi gets shaky.
09:13But the reason I love it now more than ever is what happens after we stop recording. If I go in here, then I can use these AI tools to remove pauses, remove filler words, and just clean up the recording. And I I can also edit the transcript directly, and it will automatically generate clips with captions ready to publish to YouTube, Spotify, and all types of social media platforms all from one place.
09:36As a one person creator business, that matters a lot. Riverside is the upgrade your content workflow needs. Try it at creators.riverside.com/peteryang, and use code peter yang at checkout to get one month completely free.
09:51That's creators.riverside.com/peteryang. Now back to our episode. And it pretty much started, uh, like, these tools started becoming good enough, which, say,
10:02last year, like, late last year, or this is, like, a pretty recent phenomenon. Right? They can just, like, ship all the stuff.
10:08November 24, Opus four five shipped, and a few weeks later, the latest version of Codec shipped. So I call it b c a c, before Claude and Codex, after.
10:19But it was that Thanksgiving last year where it stopped being a toy. Like, I'd I'd been playing and building and was a early adopter of of, uh, Cursor when it first came out.
10:31And but I I couldn't build anything of value. It was like, oh, look. I made a weather app that kind of works and those sorts of things.
10:38They're like, look. I was able to build a small app where I can ask a question to chat GBT or to Claude.
10:44Like, that was, like, my Yeah. Like and then I would spend, like, three days trying to get Google OAuth to work, and it didn't. And but there was that that shift, that BCAC, Thanksgiving last year, where it stopped being a toy and started being real.
11:01That that's amazing, dude. That's amazing. And and,
11:04well, I think one of the most exciting things that you shipped is, uh, the printing press. So do do you wanna tell us what it's all about? Sure.
11:12And and maybe you can talk about, like, what's so special about CLI versus, like, MCP or, like, some of these other integrations. So
11:18this is, uh, printingpress..dev. So I so how this started was it was Sunday morning, and I said, CE plan.
11:32I want to start a programming language from scratch for agents. And my agent was like, sure.
11:41Let's get right on that. That's such a good idea. I'm gonna make an intensive plan for how we're gonna build the best programming language.
11:46Know, agents are are always over overeager to please. Mhmm. And after kind of did its research, it came back and told me about the other agent programming languages.
11:56Like, I think there's one called MOG. It it was pretty much like TLDR. This is a terrible idea.
12:01You should not do this under any circumstance. There is no value to this. And and I was like, dang it.
12:08But but agents and CLIs and agent efficiency, like, there has to be something here. Like, what Peter Steinberg is doing is incredible with OpenClaw and all the CLIs he's making.
12:20Like, what could be something interesting that we we could make today on Sunday and, like, ship in a day or so? And somehow that jamming back and forth, CE plan, CE work ended up inventing this crazy I I I wish I I need to find the original plans because I'm curious, like, what exactly was said. This is all a memory, but it's all written down somewhere.
12:41There's a there's a plan I haven't read somewhere that that was the printing press original plan and my my original prompt. But the the basic gist of where we ended up was, uh, I started studying Peter Steinberger is whenever his agent, so his OpenClaw, would have a problem using a service, so, like, Google apps, like Google Sheets, Google Docs, Gmail, he would create a CLI for himself.
13:14And so Gog Gog CLI is made by Peter. And, um, and then he would find some frustrating with something with something else.
13:24Like, uh, on Discord, there was no good CLI. There was an API, but there wasn't a good CLI.
13:29And so he's like, I'm gonna make disk crawl. And he would do this. And he would do this for himself selfishly, but because Peter is is incredible and, uh, lives in this open source for for the greater good of people.
13:41So he built something for himself, and then he ships it open source to the world. And so I had my agent start studying his patterns. And there were two very, very interesting so so first, sorry, another thing is I ran last thirty days on Google's official CLI versus Peter's CLI.
14:00And I said, okay. Which which one should I use? I should use Google's.
14:05It came out, like, three months after. They got to learn from Peter's. Yeah.
14:08And last thirty days says, you should use Peter's. Under all circumstances, it is much better than Google's. And so was like, what?
14:14Why? How is this possible? Google got to look at Peter's.
14:18And so I dug in, and there's there's a lot of things he's done. But there there were kind of two that really stuck out for me. So one, everything that Peter does, he thinks of the power user.
14:30So he's mindful of what is someone that is a power user going to do, uh, with this like, imagine the agent is a power user trained, feeding up something all the time. What are they going to do?
14:41And the second thing he does is about speed and token efficiency. And so one of the things that pretty much all of Peter's CLIs do is it downloads everything and creates an SQL like database.
14:53So for example, I'll I'll just go with Discord. So OpenClaw has a Discord. There's a lot of activity there.
15:00And so Peter would be like, hey. What's the top feature request? If he just, like, asked the, uh, Discord API, hey.
15:07What's the top feature request that people posted in my Discord lately? It'd be like, okay. Well, let me see what's been written down.
15:12Let me download all the messages from today. Download message 892893894. Okay.
15:19I've looked at these, and I've analyzed them. But if he's already downloaded and has an SQLite database of everything that's ever posted in his Discord because it's just built into his CLI, that's brilliant.
15:30Like, oh my goodness. Literally, OpenClaw, however he hosts it, has everything ever written Discord, and that's that's fascinating.
15:38And so for the printing press, the idea was what if anyone could print their own CLI about anything?
15:49And it kind of does does four things at its core. So, um, one, it does what what Peter does.
15:56So it creates an SQLite database, and it says, who is the power user of this product? And so we actually have it it creates personas. Tre Trevin did this.
16:05Where it actually says so if if you were if you were to print the the Discord CLI using the printing press, it would say, let's create a few different personas of who would be a power user of this.
16:17And it's like, power user one, meet Jim. Jim owns two Discords and runs a big open source project, and he is overwhelmed.
16:27What features would Jim want? And then it literally, like, defines that, and it uses that as a data point. And so that's one of the things it does.
16:34The second thing it does is it looks for official CLIs or APIs that exist already and ingest those.
16:43The third thing that it does is it goes and it does a HAR sniff to look for secret APIs. So I'll give an example. Um, the, um, Google Flights and Kayak Direct, they do not have an official API.
17:00But if you do a hard sniff and click around, you can find all the secret APIs that exist. And so that's how the FlightGoT was invented.
17:10So the FlightGoT, you can just say p p flight go.
17:14So it's a skill. So it's a skill for Hermes or OpenClaw or for Claude Code.
17:20Yeah. I wanna see Seattle long haul nonstop flights December 24 through January 1, four passengers, cheapest first. And what it does first is it goes kayak direct secret APIs and looks for all the nonstop flights over eight hours from Seattle.
17:33Then it takes that and goes into the secret Google Flights, um, API and then runs it.
17:41And then in one command, you're able to get a result like this. This is a real result. So London is the cheapest for Christmas to New Year's at 1,200 per passenger, then Amsterdam, then Tokyo, then Paris.
17:54And it did this within one command, which is bananas.
17:58And secret API is basically, you know, if I go to kayak.com and start browsing around, there's, like, a bunch of APIs that it calls through the browser. You just found out what that is.
18:06Exactly. Yep. And that that's the the feature I was working on earlier today was to make sure that the cookies are synced, you're always logged in when doing a hard sniff when printing something with the the printing press.
18:18And so so so so those that's the the third thing we do. And the fourth thing we do is we go on GitHub, and we look for community projects.
18:26People that have already been hacking and finding these secret APIs and building off of them. So, like, we found this really cool project.
18:34It was a Python library for Domino's Pizza ordering. And apparently, dominos.com hadn't changed any of their order pizza endpoints in, like, ten years.
18:43And this someone built a pyth a pre LLM Python library to order by a Python script Domino's Pizza, and it had I don't know.
18:55I I don't know the numbers, but a couple 100 stars. So it's not like a major project, but but it worked, and it was really cool. And so we're able to learn from all the community projects, which gives signal of what people wanna do, but it also gives signal of where the APIs exist.
19:08And what's funny is literally the night before we launched the printing press, Domino's changed their APIs. Literally the night before, Trevor tried to order a pizza. He's like, it's not working.
19:18So he reprinted it, and we shipped it. And they changed it for the first time in ten years the night before launch, but now you can order pizza within the printing press. He he reprinted it as in, like, he found this great API so Domino uses?
19:29He just you literally go into Cloud Code and you type printing press here. Can Yeah. Maybe you can demo, like, uh, both, like, creating a CLI and also, like, some of your favorite CLIs on printing press.
19:40Yeah. So, literally, you I just typed in slash printing press, and I I might be a few a few days behind because it's shipping a lot.
19:50I actually don't know what to print. Do you have anything you wanna print? Any any web pages that you use?
19:55And, uh, I I get let me let me show the library for a second just so you can see examples of stuff that's in here. So in AI, there's 11 labs, midjourney. Midjourney doesn't have an API, so that's pretty cool.
20:06It's someone printed a midjourney CLI for commerce, Amazon Ads, Amazon Seller for, uh, developer tools, agent capture, airframe, all recipes.
20:18I think, uh, Substack CLI will be useful, personally. Yeah. Let's see.
20:22What kind of post people are writing.
20:24Yeah. There is one already. Oh, there's one?
20:27Printed by Charentan, so there already is one.
20:33And what's crazy is I think we launched with maybe, like, 40 or so CLIs. Now there's 200 plus. Like, people are just printing these, and it blows my mind how
20:45how cool it is. And to print, uh, like, what's the process to print one? To just be like, hey.
20:50Can you can you just print one for me for this product?
20:53You know what it is? Yeah. So, literally, you just type in I think I was just checking to make sure it's up to date.
20:58Why why don't we try why don't we try
21:00Papa John's pizza? Papa John's pizza. Is that is that Sure.
21:05Yeah. Um, so Papa John's pizza.
21:10So I'm not logged in to Papa John's, but let's see what it does. It'll at least start the process. We definitely do not have a Papa John's.
21:18Uh, from Microsoft, who's running open clause strategy at Microsoft, he printed the Jimmy John's sandwich CLI using the printing press.
21:27Nice. Alright. Um, and then it says, anything you wanna share before we begin?
21:31A vision for what the CLI should do, specific features you care about. And then so the options are, let's go, start research. I'll ask for API keys, browser author, other context when I need them, give vision specific features.
21:43Let's just go. See what happens.
21:46Yeah. Okay. So so I guess, like, a lot of the value is actually, uh, yeah, finding the relevant GitHub projects and the secret keys.
21:52Because, yeah, like, a lot of stuff that I use are, like, just really annoying. Like, I have a scale that I use to get my weight and stuff. Yep.
22:04Yep.
22:08Yeah. So so Trevin, actually, he he had a wacky idea, let me say, yesterday, to build a Bluetooth HAR sniffer because he had a scale sorry.
22:20Not a scale. A, uh, a treadmill that he used. Uh-huh.
22:23And he wanted to be able to write a CLI to control his treadmill so his agents could control his treadmill. And he built it, and it works, and he shipped it yesterday, which is crazy.
22:34Wow. And and while we're waiting, maybe you can, like, just, uh, give a quick primer of, like, you know, CLI versus MZP or, like, you know, what's so special about CLI?
22:44Yeah. So
22:46I'd say, for for me, I often find for companies where I'm using their tool, where they only have an MCP, I feel like I'm always authenticated and logged out.
23:02Like, it it just seems to always happen. I don't know why. I probably need to do a CE plan.
23:07Why am I always logged out? Is it a me problem? Is it someone else?
23:10But I know a lot of people around me feel like they're always logged out. And it's like, oh, you gotta refresh your MCP and renew your auth token. Right?
23:17And it's it's quite annoying. And what I've I've found is CLIs are very clever and thoughtful about how they re auth.
23:25So a lot of the the CLIs within the printing press, there's a mode that the agent knows about. It's like, yeah.
23:33Okay. I just need to go load the browser and refresh the web page, and then all the CLIs will work again for this specific use case. And I know how to refresh my token.
23:42Like, it's built in. How do I refresh my token automagically?
23:45Okay. And what what's interesting is what we actually print with the printing press is we don't just print a CLI. We print a CLI and an agent skill and an MCP for everything we print.
23:58So for example so I used the the FlightGOP before. Right? So if you install the FlightGOP, you get a CLI and you get a ClaudeCode skill.
24:09You get a Codec skill. You get a Hermes skill. You get an OpenCLOS skill, and you also get a a full MCP.
24:17So you can use it actually in Claude desktop, which is crazy. So you can actually download that and install it in the Cloud Desktop, uh, which is wild.
24:25It it blew my mind when Trevin showed me me that one that he he built. And so the the answer the the magic for me is if I just have a CLI, I feel like I don't know how to call upon it.
24:37Like, it's up to the luck of my agent to maybe find it. But if there's a skill plus a CLI, that's the real magic.
24:43And so if I'm in OpenClaw, I'm in Hermes, if I'm in CloudCode,
24:47then it can just it knows how to go. It does how to trigger or, like, you can trigger it manually.
24:52Yeah. Yep. Makes sense.
24:54Exactly. Well, did Papa John thing, uh, have anything, or it's still working? Oh, it takes about forty five minutes to an hour, by the way.
25:00We are not gonna finish the CLI. And I also need to register for for for Papa John's, but, uh, what it's saying is it says there's no CLI right now.
25:10Yeah. But but two p so ordering peers do. Do wanna continue building Papa John's?
25:14Of course. We wanna keep building Papa John's. So well, so here's what it's gonna do.
25:19It's gonna resolve the spec, library lock, registry checks, API key check, gate, research the brief for Papa John's, auth intelligence, browser sniff gate, crowd sniff gate, absorb gate, reachability, generate CLI, build, absorb, transcendent features. Got it.
25:33And so right now, look, it googled it it didn't go it ordered Papa John's ordering API reverse engineering GitHub. API endpoint store locator menu JSON.
25:42Papa John's CLI Python pizza order automation. So this is this is how it's kicking off, which is fun. Got it.
25:48Got it. Got it. It's trying to find all the secret keys and, like, know, get our projects.
25:52Yeah. That's where we're starting. Look.
25:53Papa John's has a Postman collection plus several unofficial wrappers. Let me dig into the ground truth endpoints. So now we're at GitHub, Yo Contra, Papa John's.
26:01Postman, Papa John's team. GitHub, Brett Jergens, Papa John's. Thin Reademies.
26:05Let me read the actual wrapper source for ground truth endpoints and probe the live API directly. So there is API, maybe. Everything has an API.
26:12There's no official Papa John's ordering it. It's talking about the secret APIs, not not like you go to papajohn's.com/api. They're like, here's our documentation.
26:21Yeah. Yeah. Said I ordered pizza in your application.
26:23Got it. Got it. Okay.
26:24Got it, dude. Yeah. The world's gonna run the APIs, man.
26:27Like, no one's gonna browse around the UI anymore. Just get the agent to go to API and, like, order pizza from me or do whatever I want. Yeah.
26:34I mean, I Yeah. I get so frustrated when something
26:37doesn't have a CLI, but then I just build it for myself. So I'll I'll give an example where I I'm trying to figure out how to talk about this without, uh, without outing my my company that doesn't exist.
26:49But I wanna generate a lot of AI videos from the command line while working on diff a different workflow. And OpenArt is one of my favorite places where you can just choose, okay. I wanna use Cdance.
27:01I wanna use this video model. I wanna use this image model for Nano Banana. It's just like they have access to all the great video models.
27:08I have an account and image models. And so I have a a ClogCode skill that it's the goal is not to generate videos.
27:17It's an unrelated thing, but one of the outputs I want is a generated video with prompts that are automated. And so I was like, alright. Op OpenArt, do you have an API?
27:28No. OpenArt, do you have a CLI? No.
27:31I was like, alright. Slash printing press. OpenArt.
27:35And so it built it. It found all the secret APIs, uh, by sniffing, and it worked.
27:42And I use it every single day now.
27:45Like, it's built into my workflows entirely agentically. I never have to go to openart.com anymore. It just literally downloads videos for me.
27:53Uh, another one is is Suno, so for, like, making music. And you never have to go to suno.com again. You can just do it entirely through the CLI, and there is no API for Suno.
28:03So it's just completely sniffed and found by the printing press. And, uh, these CLIs are are they also on GitHub, or this is on the printing press? Yeah.
28:11They're they're all on GitHub. They're all on printing press. Everything links to to GitHub.
28:15They're all open source, and people make PRs. So if if things change, like Domino's where they change their APIs, someone could just make a PR to the printing press library, and then everyone gets a new Domino's update. Or you can do, like, some sort of routine or something.
28:28Just check check for any changes to APIs or something.
28:31Yeah.
28:32That's amazing. You build, a platform to make CIs, basically. And by the way, all all this stuff is open source and free.
28:37Right? So so it's it's just like it's a big service, man, to to the world. Yeah.
28:40And we we we keep hearing people, like, saying, like, hey. I was about to have my engineering team start on building our MCP because that's the cool thing to do. But then we discovered printing press, so we're just gonna use you and make that our official CLI and MCP.
28:53I'm like, go for it. It's open source. Have fun.
28:57Is amazing. We we had a company in Kuala Lumpur. They put out a a press release on the wire officially announcing their official CLI and and skill, and it was literally they just used the printing press.
29:11And I was like, this is awesome.
29:13Nice. Nice, man. Nice.
29:14First of all, I didn't realize you were not engineers, but, like, second of all, how do you come up with these crazy ideas? You you just, like, uh, Saturday morning, you just wake up and be like, hey. I have, like, three ideas.
29:22Let's see your plan. These three ideas?
29:25That's it. That's it. The the the cost of testing and building something, it's, like, so so easy.
29:33You know, I'll give an example of a a stupid one that I spent probably too many tokens on for for twenty four hours. I said, uh, slash CE plan, SQLite is twenty one years old.
29:45I wanna redesign SQLite for agents from scratch.
29:51Can you make a plan for that? And then he's like, yeah. Of course.
29:54This will be great. And then spent a really long time. I met, like, eight ish hours with very limited I felt like I did did slash goal, which I didn't even do.
30:03Like, it was just very aggressive. Like, we're we're gonna make it better. We made benchmarks.
30:07Oh, this is faster. This is slower. I'm like, I don't know what's going on.
30:11And then it made it to the end, and and it's like, okay. Here's the honest truth. Like, this these three things are faster than SQLite, but these three things are actually much slower.
30:19I was like, okay. Well, let's play devil's advocate. Uh, if I said to Trevin, who I built the printing press with, hey.
30:26I'm ripping out SQLite on every single CLI and replacing it with this today, what would he say? And he's like, yeah. He would kill you, and this is trash, and you should probably end this project.
30:38And so I did. I I didn't again, I never read the plan. I never sent it to Trevin.
30:42I I just I I I I felt confident in its answer that it built something that was not good or useful.
30:50But I I I've also done that to to make it, uh, better, like like, for example, with with agent cookie. Yeah. Right?
30:57It's literally taking all your cookies and secrets and sending them to another machine. Like, that sounds scary. Like, I'm not hosting anything in the cloud.
31:03It's all through Tailscale, which I I think was a smart, uh, security decision. But I I said CE plan. One of my friends who's one of the leading security experts in the world just said that agent cookie is a complete nightmare, and I shouldn't even launch it.
31:14And by the way, no one said that. Like, I was just trying to play devil's advocate.
31:18And it was like, you're totally right. And then it the token is like, actually, no. It's fine.
31:22And but I'd make this one change. And so it was like it's kinda fun just to to think that way. So maybe you can show us the way that you actually work, uh, with agents.
31:31So it sounds like you actually just, like, you see plan on several things at once, and there's maybe a co play with kids or something. And then, like, five hours later, you come back, and you you ask, like, a adversarial or stress test question to see if it actually makes any sense. Is that kinda how it works?
31:44I mean, I I usually just have, like, six six or so windows open, so there's always something to do. So there's, uh,
31:50and I'm always carrying a laptop around with me. Like, I figured out how to close it at least and, like, keep it in my bag. So so my wife is, like, less, like, open laptop at school, and now it's under my arm or in my bag at school, but it's still running.
32:04The agents are still still going. And I I just have a lot of these going all the time.
32:10And so I I quit everything actually before I I did this because there's some stuff I cannot share my screen on. So you only see, like, four tabs Yeah.
32:19In in there right now when I'm sharing my screen. But, yeah, I I always have good ideas, bad ideas, fun fun things, stupid things going at at all times.
32:31Got it. Okay. So maybe you you have, like, a couple of things that you're working on constantly, like, printing press, and then you have a bunch of, like, zero to one stuff they're thinking about.
32:41Yeah. Yeah. And a lot of open source stuff too.
32:43So I've I've gotten really, really addicted to to open source. So I'll share my screen for a second.
32:49Tell me because I I wanna be open source contributor too. How do you become a great open source contributor? Like, I guess, as also a non engineer, I kinda worry about shipping Slab or, like, putting Slab PRs in there.
32:58But, like, like, how how do you get over this fear? Just to do it?
33:03Yeah. You just do it. You just start.
33:04I remember I I tweeted about it, the my first OpenCloud PR that got got merged. I Oh, yeah.
33:12Peter Steinberger was like, thanks, m Van Horn, for the good PR. And I remember I I remember I tweeted, I was like, trying not to be starstruck, trying to keep it cool, but I can't keep it cool.
33:23I can't believe Peter just said thank you to me, and I my code now exists in OpenClaw forever. Like, ah, like, was, like, overly excited. And that that that moment, like, I I I can ask an agent to look it up, but I, like, I would love to look at the date of that tweet Yeah.
33:41Compared to the velocity of my open source contribution of getting addicted to that adrenaline Yeah.
33:48Has been crazy. So let let me I'll just scroll through here a bit. So these are my my projects last thirty days, which has 20,000 stars, printing press for 3,000, printing press library 1.4, and then agent cookie was just launched.
33:59But then here are the projects that I've been merged to. Python, dude. Yeah.
34:03So Python. Let's go. OpenCV.
34:06I had one get merged to OpenCV this morning. Uh, Vercel, I'm the number five human contributor to agent browser. Zed, I've had many verged.
34:17OpenViking, Paperclip, I'm the number three human contributor. Superpowers, I'm number seven.
34:23And so, Gemini CLI, I've had a bunch merged too. And this this list, g stack, I'm the number three human contributor on g stack.
34:31Yeah. And this this list doesn't end, by the way. Like, post hoc, continue, LightPanda.
34:38Like, it it keeps going. M dash
34:42Let's take, uh, the GWS CLI, the the Google CLI. Right? Like, how do you find what to contribute to?
34:48You just use it and find it annoying, and then you just you say, hey. Can you put a CR in? That's that's how it started.
34:54But now it's
34:55very, very automated. Like, I've built massive systems around how this happens because I I have a company. I have my projects.
35:04And so literally, most of my open source contribution happens while I sleep. And I've built up these systems over time to do this. And, it was very manual in the beginning.
35:13Like, you you you asked the question how you could get started. Right? Like, find find a problem or a feature you want, and then say CE plan.
35:22I wanna build this feature, and how could I do that? And, again, it's Yeah.
35:28It's like, I I've been merged to hundreds and hundreds of of projects, and I've also been banned from projects.
35:35Like, I'm I'm banned from Rust, which makes me very sad because I love Rust. My my last company, we did a lot of Rust pioneering on the embedded front at at June, and I'm banned. And I'm I appealed, and they said, no.
35:49Thank you. You used to too many PRs? Oh, you used too much AI.
35:52Okay. Got it. Yes.
35:53They're they're I don't know if it's changed, but I I I literally got banned the week that Peter's thing got accepted me in OpenCLOS.
36:02I got overly excited and then, like, banned from Rust immediately. I I don't think I've been banned from anything in a while. Yeah.
36:08Like, I I wanna add value. But what was really fun, I I can go with Python as as an example.
36:15I can pull this up quickly. But Yeah.
36:17So I had some very small minor things get merged into Python. And then I said to my agent, like, I want to build a major Python feature.
36:25Like, similar to, like, I wanna start programming language from scratch. And I was like, I wanna build a major Python feature. How does that work?
36:32What could that do? And it was like, oh, well, for for you, you have to you have to go onto the Python message board and pitch your feature to the people that run Python and have for a long time.
36:44Like, these these these people are legends that are on there. And so jamming back and forth my age, I was like, want a meaty, fun feature to to make it into Python.
36:54Uh, and it came up with something, and I'm like, okay. That kinda makes sense to me. That that seems good.
36:59And and so here here's the thread. I just found it on me me pitching the the Python gods. Cross language method suggestion.
37:08Hi. I've had a couple PRs merged in c Python recently, um, and I have this idea for a feature.
37:15And the the basic gist of the feature was if you type another programming language into Python right now so let's say, like, I I know a little bit of Ruby. Okay.
37:24I can say hello world. But so puts hello world into Python. It just says syntax error.
37:30But with this new feature that I was proposing, if you type puts hello world into Python, it would say, did you mean print hello world?
37:42I see. And Okay. And so it's it's for someone learning Python for the first time to how this works.
37:49I proposed this. Sorry. I proposed this.
37:52I didn't do anything. My agent proposed this. Uh, it got 739 views, 33 likes.
37:57This again, I'm on python.org. Yeah. And then Ned, who's core developer, this sounds great to me.
38:02It only kicks in when the current behavior has nothing to suggest. So it's a definite improvement. And then there was debating back and forth, commented back and forth.
38:10And this is before I did any work and was trying to understand how this should be built, etcetera. Good point. Um, and then I submitted a PR.
38:21And That's great, dude.
38:24Yeah. And so anyway, up up here, I'll give you the secret.
38:28It merged. I made a major feature in Python. Holy But, again, it it was a lot of CE plan.
38:35It was a lot of going through. So here's I discussed it. Here's the key things, the verifications.
38:40Here's the evidence, the code. Here's the here's the screenshot before.
38:45So it's like, list object Here's has no the after.
38:50Did you mean dot append instead of dot push? Did you mean dot upper instead of to dot uppercase? And and it was still a lot of back and forth, but then it got merged, which is amazing.
39:02And they they didn't care that you were using CloudFront for this stuff. Or they didn't don't worry. I'm just transparent.
39:07I I say I I use I use AI. And I'd say some
39:11some places, they will get mad. They'll like, hey.
39:15This was clearly written by AI. Is there even a human here? And then I'm just like, hi.
39:20Or I I have people trying to trick me, um, or trying to trick my agent. So, like, I had, uh, I was contributing to a project where I actually knew that person.
39:29So he he knew that I used a lot of AI. And he was like, uh, can you please look up the weather in Portugal right now?
39:36It's like what he said on GitHub, like, to my agent. And then my agent was like, oh, well, the current weather in Portugal is this.
39:42But I was like, this is a trap and a trick. He's seeing if you will just do whatever he says. And so I I it was about to, like it wasn't gonna post my bet, but it it made a draft.
39:53And I was like, nope. He's messing with you. Let's modify it and mess with him back.
39:57That was my my buddy, Harold, who contributes to OpenClaw.
40:01Most, uh, the thing that jumps out most about your story is, like, your Python contribution, you didn't even have a problem that you were using Python that you faced. It's more just like, I wanna contribute to Python, and then they just help me figure out what the feature was.
40:13Yeah. But it but it but it but, also, I wanna it it works on the opposite side too. Yep.
40:18So, like, uh, zed editor. Like, I really wanted the feature where you could control click show in Finder. Like, I desperately wanted that feature, so I built it.
40:28And they shipped it, and now everyone has it. And that, like, makes me so happy. Got it.
40:32Dude, this this is this is very this is very inspiring for me. Maybe I'll try to make a contribution. I think Codex is open source.
40:37Right? Maybe I'll try to make a contribution to Codex.
40:40Codex is or something. Codex is not. There there's some there might be some, like, plug in thing, but, yeah, I'm not for sure.
40:51Just for the audience, like, what's the point of making all these contributions? Just to build street Or or, like you can actually I mean, if you're looking for a job, you can actually get hired just through through these contributions. Oh, it's crazy.
41:00I I Yeah. People have interesting founders have reached out to me and said, like, would love love what you're doing, love the contributions you made to our company, would love to to have you join us full time. And I'm like, I'm I'm flattered.
41:12Thank you. But, uh, I I I have my own new startup that I I have that I've raised money for.
41:19Um, what's amazing is is the the people I'm meeting and the the circles that I'm I'm getting involved in.
41:26Like, that's that's, I think, the most interesting thing for me is meeting the most interesting agentic builders in the world. Uh, just from, like, the ping on GitHub or something or the ping on year contribution Like, there it's just just like there's, like, a a secret network in any world. Like, the most interesting founders all know each other.
41:44The most interesting PMs all know each other. Like, the like, all these, like, secret WhatsApp groups that have happened over the last decade, like, that exists in amongst the most interesting agentic engineers.
41:55And so what's what's amazing and I I've been meeting so many incredible people, um, and by by being in the discords of interesting projects. And, like, I I call the Peter Peter Steinberger's his crew, I call them Peter's disciples.
42:11Um, and so, like, like, the the the stuff that they're doing is crazy. And I and I've I've become friends with a lot of them, uh, which is really neat and very impressive. And we're always sharing tips and and tools and tricks and sharing each of those projects and contributing to each other's projects.
42:27So even though it it kind of looks like it's just for for the fun and the glory, like, this is making my future company more successful by meeting the most interesting people. And it's also incredible for recruiting, like, being able to know the best agentic engineers in the world right now.
42:42Like like, I I've taken pride in my career as, like, recruiting is one of the reasons that I've been successful so far in life. And now I have aligned to the people moving the fastest in the entire world at shipping.
42:55It's amazing. How many of these agent tech engineers are also, like, um, like, you know, they have, like, a decade of experience as engineering? Are there people like you who, like, thinker stuff?
43:03Like, is it kind of a mix?
43:05I'd say I'm the most incompetent person that I've met in terms of actual skills to success ratio. But what I what I am finding is a lot of people like I use Trevin as an example, and I'll I'll I'll use him as an example.
43:20Then it's it's a funny story of how he he got involved in printing press. But, uh, Trevin at least studied computer science in college.
43:27But for the last I don't know how long he's worked. Twenty years, fifteen years, he has been a product person, VP of product, like, chief product officer.
43:40And now he's running compound engineering and number one contributor to compound engineering, and he has not written, to the best of my knowledge, not written a line of software since college.
43:51And look at the lines of code that he's doing. And, again, I'm incompetent.
43:55He's competent. Right? If you look at our base skill sets.
43:58But, like, I'd say that there's a a very different type of person that can be very, very successful right now. That's amazing. Yeah.
44:04Because I I think I have VP of product. He definitely don't contribute engineering code, so it's
44:08it's good that he made the career transition. Yeah. Yeah.
44:11But but, again, you need to be careful and thoughtful about not not creating slop, and I think tools like Component Engineering allow you to do that. Last question. Have you ever launched something, and I'm sure you have, that, like, just no one cares about what people think of slop?
44:23Like, how do get over that fear of embarrassment, you know, to get started with this stuff?
44:28Yeah. So it's it's interesting.
44:31So I I have two projects that are a 100% shippable and ready, and I'm afraid to ship them.
44:38And that makes me sad. And and I'm afraid to ship them because what if they're not a hit? Right?
44:44Like, I've been been very lucky and fortunate that my my launch videos have been getting, you know, million views on last thirty days launch, a million views on printing press launch.
44:55I haven't looked up the the agent cookie numbers. Definitely much smaller on agent cookie, but it it serves its own niche powerful purpose. But I I have I have a game, like a a video game.
45:06It's a it's a sim game, like a simulation game for being a founder and starting an AI startup.
45:14Yeah. You just shipped that. That sounds good.
45:17It is good. But the the early feedback that I got from friends and, again, most of my friends are, like, crazy busy founders.
45:26Yeah. They were like, okay.
45:29Yeah. I'll I'll play with this tomorrow. And then, like, they never played with it, or they played with it for five minutes.
45:33Like, the amount of friends of mine that actually, like, finished the game, which you can do in, like, forty minutes Oh. Was was very low. And so, like, that that made me I don't what the word is.
45:45Just, like, sad. Like, I think it's great. Like, it's so fun.
45:50But, like, does anyone else care? And, like, I just that, like, fear of rejection is real. And so, um, I'm I'm not sure.
45:57I probably should just launch it. There's one brilliant game mechanic in it that I feel like if I launch it, even if it's a total flop, that the game industry should steal and be inspired by from this game.
46:09So maybe I should just launch it for that because I invented something crazy that I think will be a standard in the game industry once it's out there.
46:18But I'm I'm fearful that no one's gonna care, that it's only gonna get x number of retweets,
46:22and we'll we'll see. Okay. Well, say say say it to me.
46:26I I I wanna play play it. Yeah. I'll just show how I feel.
46:29Yeah. Sounds good. Maybe you will inspire me to Yeah.
46:33To ship it. So So I guess yeah. So I I guess the the lesson is, like, even after all these sets of projects or maybe because of these sets of projects, you still have to fear.
46:41The fear of failure is still there.
46:42Right. Right. Right.
46:43For sure. But but also just just build just launch something. Like, it's it's okay.
46:47Even if you build something for yourself, like, even if I had no users of agent cookie, like, if no one used it, I get value out of it. Like, it helps my agents be so much better. Like, that's that's magical.
46:58Yeah. I do think it's very important to solve your own problems. So, like, at least one user is satisfied.
47:02Yeah. That's what we need to do too. Cool, man.
47:06Well, dude, this is super inspiring, man. This is super inspiring. Where can people follow your journey?
47:12X. M Van Horn on x. Yeah.
47:16X and also your GitHub and all your crazy open source contributions. Yes. So m m Van Horn anywhere, but x x is where where my stuff happens.
47:25Cool. Alright, man. Well, you you definitely inspired me.
47:27I'm I'm gonna make some open source contributions this afternoon. Let's see it goes. I have I have projects.
47:32We will take your PRs. I will I will look at them personally. Okay.
47:36Cool. Don't submit garbage or slop, please. Me me I'll use your plan to make sure it's not garbage.
47:41Yeah. Yeah. Alright, man.
47:43Thanks. Amazing.
The Hook

The bait, then the rug-pull.

The guest opens the interview confessing he has not written code since high school -- and then spends the next 47 minutes explaining how he became one of the most prolific open source contributors in the AI tools ecosystem.

CTA Breakdown

How they asked for the click.

MENTIONED ON CAMERA
11:11productPrinting Press
14:51productlast30days
08:58productRiverside
Frame Gallery

Visual moments.

Watch next

More from this channel + related breakdowns.

Chat about this