Live data from Hacker News

Ex-GitHub CEO launches a new developer platform for AI agents

entire.io

591–600 of 625 posts

Re: Ex-GitHub CEO launches a new developer platform for AI agents

#591
post #504

Checkpoints are not granular enough. You have to grab a work summary before the context window closes, not when you decide to commit. I have thousands of work summaries from one project alone. I use them to traverse sessions. This is the agent that does the work: https://github.com/ChicagoDave/devarch/blob/main/docs/.claud... My success with Claude Code is directly related to this agent and its output.

interesting. Do you have an example of what exactly it outputs and how you store that? how do you use the outputs? As a pointer to a new agent with fresh context to get them up to speed?

This is the current context docs in my text adventure authoring platform Sharpee:

https://github.com/ChicagoDave/sharpee/tree/main/docs/contex...

Note I archive these to a different non-repo folder regularly. It's fun to tell Claude to go through them like a RAG implementation and note the progress of my development over time, which I update semi-regularly at: https://david.cornelson.net/sharpee-status-20260109.html

Now updated to https://david.cornelson.net/sharpee-status-20260211.html

Re: Ex-GitHub CEO launches a new developer platform for AI agents

#592
post #254

Earlier quoted context omitted.

Sure... you `git add` the context text generated by AI and `git commit` it, could be useful. Is that worth 60 million?

It’s good to know that a few decades later the same generic Dropbox-weekend take can be made.

> It’s good to know that a few decades later the same generic Dropbox-weekend take can be made.

The dropbox-weekend take wasn't made by the intended target for the product.

This is.

Re: Ex-GitHub CEO launches a new developer platform for AI agents

#593

Earlier quoted context omitted.

@dang with the launch of open claw I have seen so much more LLM slop comments. I know meta comments like mine aren't usually encouraged, but I think we need to do something about this as a community. Is there anything we can do? (either ban or at least requiring full disclosure for bot comments would be nice). EDIT: I suspect the current "solution" is to just downvote (which I do!), but I think people who don't chat…

The moltbook-ification of every online forum seems inevitable this year. I wish we had a counter to this.

I've been thinking about this, one solution I wonder if to put a really hard problem in the sigh up flow that humans couldn't solve, if it's solve in the signup, it's a bot, not sure how tf to actually basically captchas flip, however I suspect this would only work for so long.

Re: Ex-GitHub CEO launches a new developer platform for AI agents

#594

> Spec-driven development is becoming the primary driver of code generation. This sounds like my current "phase" of AI coding. I have had so many project ideas for years that I can just spec out, everything I've thought about, all the little ideas and details, things I only had time to think about, never implement. I then feed it to Claude, and watch it meet my every specification, I can then test it, note any bugs,…

I’m testing this as well, but I’ve no idea people are getting good results out of Claude. What a waste of money… it’s incredible. It will get stuck on everything: basic string substitution, trying to install Python dependencies in a Node project, finding usages of something it’s trying to edit, and on and on.

Re: Ex-GitHub CEO launches a new developer platform for AI agents

#595
post #504

Earlier quoted context omitted.

interesting. Do you have an example of what exactly it outputs and how you store that? how do you use the outputs? As a pointer to a new agent with fresh context to get them up to speed?

This is the current context docs in my text adventure authoring platform Sharpee: https://github.com/ChicagoDave/sharpee/tree/main/docs/contex... Note I archive these to a different non-repo folder regularly. It's fun to tell Claude to go through them like a RAG implementation and note the progress of my development over time, which I update semi-regularly at: https://david.cornelson.net/sharpee-status-20260109.html…

60 million dollar idea and you’re giving it away for free!

Re: Ex-GitHub CEO launches a new developer platform for AI agents

#597

Earlier quoted context omitted.

Of course it's how it works, how else do you justify a company that is making negative profit into some how being worth $300M? Like that's just the game, IDK why people accept it. It's not democratic and all it does it prime the population for fraud and abuse.

Yeah, that's not the game at all. Have you ever invested in startups?

I don't need to "play the game" to realize private valuations are just marketing fluff not based in reality. It's literally "the greater fool" theory in action. When the incentives are to just lie and not put out something with some actual scrutiny like a 409A, it's quite clear what game is being played.

But yes, I would totally love to invest in startups with people's pension funds. It seems like the perfect scam where the only losers are the public that allows such actions.

Re: Ex-GitHub CEO launches a new developer platform for AI agents

#598

> The game has changed. The system is cracking. Just say what your thing does. Or, better yet, show it to me in under 60 seconds. Web sites are the new banner ads and headings like that are the new ` `.

It’s also a completely meaningless headline. It could be the beginning of just about any Medium post, corporate blog, or cyberpunk Steam game.

Re: Ex-GitHub CEO launches a new developer platform for AI agents

#599

The product isn't the software, it's your data. The VC's didn't give Dohmke $60m to build a new frontend for Git/ GitHub... They're going to capture your conversations and code with AI (The outcomes) and use that to train better models which they'll rent back at you.

100%, and many developers are gonna fall for it. I’m off GitHub because I don’t think every corporation under the sun should profit off my work.

Re: Ex-GitHub CEO launches a new developer platform for AI agents

#600
post #229

Earlier quoted context omitted.

I'm pretty much just rawdogging Claude Code and opencode and I haven't bothered setting up skills or MCPs except for one that talks to Jira and Confluence. I just don't see the point when I'm perfectly capable of writing a detailed prompt with all my expectations. The problem is that so many of these things are AI instructing AI and my trust rating for vibe coded tools is zero. It's become a point of pride for the hu…

I think the rationale is that with the right tools you can move much faster, and not burn everything to the ground, than just rawdogging Claude. If you haven't bothered setting up extra tools you may still be faster / better than old you, but not better than the you that could be. I'm not preaching, that's just the idea. > That is not software engineering or development, it's brogrammer trash. Yes, but it's working.…

You're reading it though, and I imagine you are applying comprehension to it based on your experience. It's not vibe-coding any more at that point, I'd call it rapid application development. That's what Rails did in, what, 2010? Maybe earlier? Except it was generated scaffolding code created through reflection and not machine learning.

It's when you take yourself out of the loop and trust the process that it goes in the wrong direction.

Post reply on HN