Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

When AI Builds Itself: Our progress toward recursive self-improvement

anthropic.com

Post reply on HN
Print view View original

271–280 of 738 posts

Re: When AI Builds Itself: Our progress toward recursive self-improvement

#271

Post by 0xbadcafebee »

aleqs wrote: post #40

Okay, so anthropic has amazing AI which supposedly writes most of their code and can continuously improve... meanwhile they have outages on a regular basis, and any kind of long-running work will now consistently hit 'API Error: Server is temporarily limiting requests'. Not sure of this is intentional to force a reduction of token usage, but at this point I need to build around these throttling limits and outages wit…

Have you considered just... using OpenAI? They are more reliable, models are just as good, and their subscriptions provide more requests per dollar.

Re: When AI Builds Itself: Our progress toward recursive self-improvement

#272

Post by overgard »

oersted wrote: post #223
javcasas wrote: post #215

Earlier quoted context omitted.

Nope. Used to render on the terminal like a game engine. https://x.com/trq212/status/2014051501786931427

This kind of immediate-mode rendering is quite standard for TUIs. Although immediate-mode rendering tends to be significantly simpler and use less memory than retained-mode rendering, at the cost of some redundant computation. So I am not sure if this is the reason for the bloat. It’s possible that it doesn’t play well with JS garbage collection, since it recreates the whole UI structure for every frame (which tends…

I haven't tried to make a TUI admittedly, but double buffering is the oldest technique on the planet. A TUI doesn't even need to pay the cost of a lot of pixels since its effective resolution is much lower

Re: When AI Builds Itself: Our progress toward recursive self-improvement

#273

Post by fg137 »

malfist wrote: post #104
torben-friis wrote: post #36

>A caveat: Lines of code is an imperfect measure, as it measures quantity over quality. So 8× lines of code/engineer/day in the second quarter of 2026 is almost certainly an overstatement of the true productivity gain. Nonetheless, it indicates an acceleration. At Anthropic, we don’t reward people for how many lines of code they write; rather, team members are producing more code simply because they’re using AI syste…

One of my co-workers just asked me to review his pull request that was all AI generated. 600 files were touched, over 40k lines of code added. I'm sure he thought that was a crowning achievement, proof that AI can enable 10X developers, after all, what engineer could write 40k lines of code in a week? I declined to review it, stating that I couldn't possibly vet 40k lines of code, and wouldn't put my reputation on th…

Same here. A co-worker touched a few hundred files in a PR and asked us to review. They merged it directly to main when nobody approved it. (The repo was not set up to enforce PR approval.)

I don't personally use that feature, and I couldn't care less at this point. If our customers are frustrated by the bugs, at least my name is not on it.

Re: When AI Builds Itself: Our progress toward recursive self-improvement

#274

Post by AceJohnny2 »

javcasas wrote: post #188
airstrike wrote: post #112

Earlier quoted context omitted.

This might explain it, in the opposite way it was meant to: https://fxtwitter.com/trq212/status/2014051501786931427 > Most people's mental model of Claude Code is that "it's just a TUI" but it should really be closer to "a small game engine".

> For each frame our pipeline constructs a scene graph with React then > -> layouts elements > -> rasterizes them to a 2d screen > -> diffs that against the previous screen > -> finally uses the diff to generate ANSI sequences to draw Yup. Overengineering.

This is a decades-old design pattern when CPU >> IO. Emacs has been doing just that since the 80s, when people were complaining about "Eight Megs And Constantly Swapping". See "redisplay" [1]

This minimizes screen flash. You can't rely on terminals doing double-buffering.

[1] https://github.com/emacs-mirror/emacs/blob/c29071587c64efb30... or a more user-friendly overview, Daniel Colascione's seminal "Buttery Smooth Emacs", snapshotted at e.g. https://gist.github.com/ghosty141/c93f21d6cd476417d4a9814eb7...

Re: When AI Builds Itself: Our progress toward recursive self-improvement

#275

Post by andriy_koval »

overgard wrote: post #266
andriy_koval wrote: post #209

Earlier quoted context omitted.

Maybe that gigabyte is occupied by useful information: traces/memory?

A gigabyte is a lot of memory. Even the largest context windows are a small fraction of that with any sane engineering discipline.

For each LLM interaction they likely have bunch of thoughts traces, tool calls, etc, which don't go to context, but still can be retrieved.

But I obviously don't know for sure.

Re: When AI Builds Itself: Our progress toward recursive self-improvement

#276

Post by 0xbadcafebee »

airstrike wrote: post #112
f311a wrote: post #88

Earlier quoted context omitted.

Infrastructure is a much harder problem. They can't even improve Claude Code, which eats 1GB+ of RAM. Meanwhile, my editor only consumes 80MB of RAM.

This might explain it, in the opposite way it was meant to: https://fxtwitter.com/trq212/status/2014051501786931427 > Most people's mental model of Claude Code is that "it's just a TUI" but it should really be closer to "a small game engine".

If they used an actual game engine to render a 3D UI from scratch it would be more efficient

Re: When AI Builds Itself: Our progress toward recursive self-improvement

#277

Post by keeda »

torben-friis wrote: post #36

>A caveat: Lines of code is an imperfect measure, as it measures quantity over quality. So 8× lines of code/engineer/day in the second quarter of 2026 is almost certainly an overstatement of the true productivity gain. Nonetheless, it indicates an acceleration. At Anthropic, we don’t reward people for how many lines of code they write; rather, team members are producing more code simply because they’re using AI syste…

So the more rigorous studies about AI-assisted coding productivity addressed this by keeping in place all other software development processes, including the same code review and quality standards, and only measuring throughput (PRs, LoC) before and after AI was allowed.

Hence the intepretation of this 8x number depends on whether (or how much) Anthropic engineers have changed their quality standards and development processes. They don't tell us, and I am not aware of any other indications we could use to make a judgment.

However, we can still do some theorycrafting! I'm convinced that to fully realize the potential of AI-assisted coding we need to revamp all the dev processes, especially how we validate code, and it would be foolish of Anthropic not to do so (unless they were conducting a rigorous study, which they don't claim to have done.)

My hypothesis on the future of software validation is nothing fancy, we simply want much, much more automation for tests, observability and other bespoke verification methods than we traditionally had. But then validation code will also contribute to the LoC! My observation so far of personal as well as some "vibe-coded" open-source projects is O(LoC production code) ~= O(LoC test code). So as a SWAG the upper bound could be something like a 3 - 4x speedup, which is still remarkable.

All bets are off if code quality standards are not the same.

Re: When AI Builds Itself: Our progress toward recursive self-improvement

#278

Post by malfist »

CamperBob2 wrote: post #225
malfist wrote: post #104

Earlier quoted context omitted.

One of my co-workers just asked me to review his pull request that was all AI generated. 600 files were touched, over 40k lines of code added. I'm sure he thought that was a crowning achievement, proof that AI can enable 10X developers, after all, what engineer could write 40k lines of code in a week? I declined to review it, stating that I couldn't possibly vet 40k lines of code, and wouldn't put my reputation on th…

I declined to review it, stating that I couldn't possibly vet 40k lines of code Gee, that sounds like a job for Claude if there ever was one.

You're absolutely right!

Re: When AI Builds Itself: Our progress toward recursive self-improvement

#280

Post by asdfman123 »

fg137 wrote: post #263
asdfman123 wrote: post #227

Earlier quoted context omitted.

You can do that. But I'm telling you, in tech (and enterprise shops I've worked at too) they don't care. I'm using the internal Google tools and it's helping me write code much faster too, but it still takes time. I could make the CLI tool I work on faster, but no one cares except the end users, and their minor concerns have no impact on our internal politics. At the end of the day you have to do what you're paid to…

In other words, performance is almost always an afterthought.

Sure
Post reply on HN