361–370 of 625 posts
Earlier quoted context omitted.
Survivorship bias. How many failed and commenters were right?
I'm not disagreeing, just soliciting. Does anyone have examples of products that failed in the early stages because their implementation was too trivial?
> 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.
Sure... you `git add` the context text generated by AI and `git commit` it, could be useful. Is that worth 60 million?
It's because of everybody there.
Currently no one is on Entire - the investor are betting they will be.
This guy was the ex-ceo of GitHub and can't bother to communicate his product in a single announcement post?
I am here. What did I not bother with? I wrote the blog post and it has all the details.
I don’t mean to be so presumptuous as to teach Grampa how to suck eggs, but I think Amazon’s working backwards process is instructive.
> 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.
[flagged]
> 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 built out the same thing in my own custom software forge. Every single part of the collaborative development process is memoized.
I hacked together something similar to the concept they describe a few months ago ( https://github.com/btucker/agentgit ) and then ended up not actually finding it that useful and abandoning it.
I feel like the value would be in analyzing those rich traces with another agent to extract (failure) patterns and learnings, as part of a flywheel setup. As a human I would rarely if ever want to look at this -- I don't even have time to look at the final code itself!
[flagged]
"Please don't post shallow dismissals, especially of other people's work. A good critical comment teaches us something."
This guy was the ex-ceo of GitHub and can't bother to communicate his product in a single announcement post?
[flagged]
You may not owe ex-Github-CEOs better, but you owe this community better if you're participating in it.
Earlier quoted context omitted.
The most interesting thing about everyone trying to position themselves as AI experts is the futility of it: the technology explicitly promises tomorrows models will be better then todays, which means the skill investment is deflationary: the best time to learn anything is tomorrow when a better model will be better at doing the same work - because you don't need to be (conversely if you're not good at debugging and…
the best time to learn anything is tomorrow when a better model will be better at doing the same work doesn’t that presume no value is being delivered by current models? I can understand applying this logic to building a startup that solves today’s ai shortcomings… but value delivered today is still valuable even if it becomes more effective tomorrow.
Earlier quoted context omitted.
[flagged]
Look it’s obvious at this point to anyone who is actually using the tools. We can articulate it but why should we bother when it’s so obvious. We are at an inflection point where discussion about this, even on HN, is useless until the people in the conversation are on a similar level again. Until then we have a very large gap in a bimodal distribution, and it’s fruitless to talk to the other population.
You could have someone collect and analyze a bunch of them, to look for patterns and try to improve your shared .md files, but that's about it