Live data from Hacker News

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

entire.io

581–590 of 625 posts

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

#581

Earlier quoted context omitted.

I'm not disagreeing, just soliciting. Does anyone have examples of products that failed in the early stages because their implementation was too trivial?

How exactly are we supposed to hear about something that failed in the early stages?

There are a number of ways. Obviously Dropbox would be one case of "early and didn't fail" that could have been "early and failed", and we would have heard about it.

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

#582

> Checkpoints are a new primitive that automatically captures agent context as first-class, versioned data in Git. When you commit code generated by an agent, Checkpoints capture the full session alongside the commit: the transcript, prompts, files touched, token usage, tool calls and more. This thread is extremely negative - if you can't see the value in this, I don't know what to tell you.

The most active HNers are just extremely negative on AI. I understand the impulse (you spend years honing your craft, and then something comes along and automates major portions of it) but it's driven by emotion and ego-defense and those engaged in it simply don't recognize what's motivating them. Their ego-defense is actually self-fulfilling, because they don't even try to properly learn how to leverage LLMs for coding so they give it a huge task they want it to fail on, don't properly break it into tasks, and then say "i told you it sucks" when it fails to one shot it.

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

#583

> Checkpoints are a new primitive that automatically captures agent context as first-class, versioned data in Git. When you commit code generated by an agent, Checkpoints capture the full session alongside the commit: the transcript, prompts, files touched, token usage, tool calls and more. This thread is extremely negative - if you can't see the value in this, I don't know what to tell you.

How does this differ from what Github Copilot does when writing its .github/copilot-instructions.md? That doesn't keep the transcript or prompts, but it does keep quite a bit of the context and a declarative state of the decisions/design considerations made so another AI bot can pickup and have enough context to understand the rationale. I'm also not really convinced that any AI agent wouldn't still parse the code to understand more about the context vs. just using the checkpoint.

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

#584
post #147

Earlier quoted context omitted.

> Web sites are the new banner ads and headings like that are the new ` `. It's been like this since the Dotcom era Or did you forget that you can do anything at zombo.com? It appears to be rather slow today, but here's a Wiki link for the uninitiated- https://en.wikipedia.org/wiki/Zombo.com

The domain expired a few days ago and was purchased by someone else and then changed. There's a recreation of the original here https://html5zombo.com/

Awful news to hear. May its limitless frontiers rest in peace.

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

#585

> Checkpoints are a new primitive that automatically captures agent context as first-class, versioned data in Git. When you commit code generated by an agent, Checkpoints capture the full session alongside the commit: the transcript, prompts, files touched, token usage, tool calls and more. This thread is extremely negative - if you can't see the value in this, I don't know what to tell you.

I think this is neat; in fact, I orchestrated my AI agents to do something similar (keep a log of files touched and why). And I have agents refer to the work log as well when they are unclear on why something exists.

It's not 1:1 with checkpoints, but I find such things to be useful.

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

#587
post #449

[flagged]

The world's richest man solicited investors and other people's money to buy Twitter.

https://en.wikipedia.org/wiki/Acquisition_of_Twitter_by_Elon...

I think that's just how rich people play the game. Why use your own money when you can use other people's money?

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

#589
I'm not making a judgement on the product but why is everything about selling shovels? I want to see "real" software and new innovative ideas that are developed and maintained using AI and not just shovels 'advertised' on HN every week.

Show me the next blockchain or noSQL, or git!

Everything is "I took a prompt and stapled it to another prompt and added framework".. and then. holds hand out "MONEY PLEASE!"

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

#590
post #582

> Checkpoints are a new primitive that automatically captures agent context as first-class, versioned data in Git. When you commit code generated by an agent, Checkpoints capture the full session alongside the commit: the transcript, prompts, files touched, token usage, tool calls and more. This thread is extremely negative - if you can't see the value in this, I don't know what to tell you.

The most active HNers are just extremely negative on AI. I understand the impulse (you spend years honing your craft, and then something comes along and automates major portions of it) but it's driven by emotion and ego-defense and those engaged in it simply don't recognize what's motivating them. Their ego-defense is actually self-fulfilling, because they don't even try to properly learn how to leverage LLMs for cod…

The negativity is driven by outrageous claims how AIs will replace programmers, or how english is the PL of the future. T
Post reply on HN