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

Scaling long-running autonomous coding

cursor.com

Post reply on HN
Print view View original

181–190 of 203 posts

Re: Scaling long-running autonomous coding

#181

Post by alfalfasprout »

eeL3bo1mohn7pee wrote: post #79
tehsauce wrote: post #69

I was excited to try it out so I downloaded the repo and ran the build. However there were 100+ compilation errors. So I checked the commit history on github and saw that for at least several pages back all recent commits had failed in the CI. It was not clear which commit I should pick to get the semi-working version advertised. I started looking in the Cargo.toml to at least get an idea how the project was construc…

Of 63295 workflow runs, apparently only 1426 have been successful. It's hard to avoid the impression that this is an unverified pile of slop that may have actually never worked. The CI process certainly hasn't succeeded for the vast majority of commits. Baffling, really.

You should see the code. It's true slop. The organization makes no sense.

Re: Scaling long-running autonomous coding

#182

Post by alfalfasprout »

satvikpendem wrote: post #114
ben_w wrote: post #112

Earlier quoted context omitted.

I'm not too surprised, the way I read a lot of (not all!*) the negative comments is ~"I'm imagining having to work with this code, I'd hate it". Even though I'm fairly impressed with the work LLMs do, this has also been my experience of them… albeit with a vibe-coding** sample size of 1, done over a few days with some spare credit. The positive views are mostly from people who point out that what matters in the end i…

The problem I've had with vibe coding is akin the adage of the first 90% of the code taking 90% of the time, and the last 10% taking the other 90% of the time. The LLM can get you to 90% initially but it hits a wall unless you the user know what it's doing and outputting, but that is very difficult when you're vibe coding by its very definition, meaning that you're not looking at the code at all. And then you have to…

The problem is getting there 90% but poorly makes that last 10% much harder.

Re: Scaling long-running autonomous coding

#183

Post by neuronexmachina »

fwip wrote: post #179
wilsonzlin wrote: post #178

Earlier quoted context omitted.

Thanks for the feedback. There were some build errors which have now been resolved; the CI test that was failing was not a standard check CI, and it's now been updated. Let me know if you have any further issues. > On twitter their CEO explicitly stated that it uses a "custom js vm" which seemed particularly misleading / untrue to me. The JS engine used a custom JS VM being developed in vendor/ecma-rs as part of the…

When you say "have now been resolved" - did the AI agent resolve it autonomously, did you direct it to, or did a human do it?

Looks like Cursor Agent was at least somewhat involved: https://github.com/wilsonzlin/fastrender/commit/4cc2cb3cf0bd...

Re: Scaling long-running autonomous coding

#184

Post by embedding-shape »

neuronexmachina wrote: post #183
fwip wrote: post #179

Earlier quoted context omitted.

When you say "have now been resolved" - did the AI agent resolve it autonomously, did you direct it to, or did a human do it?

Looks like Cursor Agent was at least somewhat involved: https://github.com/wilsonzlin/fastrender/commit/4cc2cb3cf0bd...

Looks like a bunch of different users (including Google's Jules made one commit) been contributing to the codebase, and the recent "fixes" includes switching between various git users. https://gist.github.com/embedding-shapes/d09225180ea3236f180...

This to me seems to raise more questions than it answers.

Re: Scaling long-running autonomous coding

#185

Post by DauntingPear7 »

Bishonen88 wrote: post #128
physicsguy wrote: post #118

I have been trying Claude Code a lot this week. Two projects: * A small statically generated Hugo website but with some clever linking/taxonomy stuff. This was a fairly self-contained project that is now 'finished' but wouldn't hvae taken me more than a few days to code up from scratch. * A scientific simulation package, to try and do a clean refresh of an existing one which i can point at for implementation details…

Similar experience here. Did sign up for Claude Code myself this week, too, given the $10/month promo. I have experience with AI by using AWS Kiro at work and directly prompting Claude Opus for convos. After just 2 days and ~5-6 vibe coding sessions in total I got a working Life-OS-App created for my needs. - Clone of Todoist with the features that I actually use/want. Projects, Tags, due dates, quick adding with a t…

Wth are your usage limits? Are they increased? I’ll hit a usage limit in about 2-3 hours of using sonnet 4.5, and opus is a weekly limit.

Re: Scaling long-running autonomous coding

#186

Post by janstice »

xmprt wrote: post #63
blamestross wrote: post #31

Earlier quoted context omitted.

I suppose brittle code is fine if you have cursor to update and fix it. Ideal really, keeps you dependent.

To be fair, that was always the case when working with external contractors. And if agentic AI companies can capture that market, then that's still a pretty massive opportunity.

At least AI is (and unlike many contract dev shops) keen to write unit tests…

Re: Scaling long-running autonomous coding

#187

Post by snowmobile »

simonw wrote: post #3

"To test this system, we pointed it at an ambitious goal: building a web browser from scratch." I shared my LLM predictions last week, and one of them was that by 2029 "Someone will build a new browser using mainly AI-assisted coding and it won’t even be a surprise" https://simonwillison.net/2026/Jan/8/llm-predictions-for-202... and https://www.youtube.com/watch?v=lVDhQMiAbR8&t=3913s This project from Cursor is the s…

Well, it doesn't surprise me that this project is just a non-compiling clone of an existing browser. Says a lot about AI in general, don't you think? https://news.ycombinator.com/item?id=46649046

Re: Scaling long-running autonomous coding

#188

Post by snowmobile »

simonw wrote: post #154
jcfrei wrote: post #135

Earlier quoted context omitted.

Surely a smart implementation would just find the chromium source on github, do some cosmetic rewrites and strip out all none-essential features?

You'd be able to see it doing that by looking at the transcript. You could then tell it not to!

I suppose Cursor forgot to tell their AI that, before claiming that it built everything "from-scratch"

Re: Scaling long-running autonomous coding

#189

Post by mjmas »

embedding-shape wrote: post #184
neuronexmachina wrote: post #183

Earlier quoted context omitted.

Looks like Cursor Agent was at least somewhat involved: https://github.com/wilsonzlin/fastrender/commit/4cc2cb3cf0bd...

Looks like a bunch of different users (including Google's Jules made one commit) been contributing to the codebase, and the recent "fixes" includes switching between various git users. https://gist.github.com/embedding-shapes/d09225180ea3236f180... This to me seems to raise more questions than it answers.

The ones at *.ec2.internal generally mean that the git config was never set up ans it defaults to $(id -un)@$(hostname)

Re: Scaling long-running autonomous coding

#190

Post by isusmelj »

I think we are just very close to the peak of a typical Gartner hype cycle around LLMs. They are useful but overhyped. There will be more posts about fuckups that happen because people run things on autopilot and cannot keep up with reviewing AI generated code.

Do not get me wrong. I use AI all day to speed things up. But I believe that there is only a small group, maybe 5 percent or less, that actually knows how to use AI properly (I'd count myself not yet in that 5%), which I see as potentially dangerous. The other issue I see is inexperienced software engineers writing software. Although I see this as a great value add and productivity boost for prototyping, I am afraid of the “I do not know much about coding but can also make PRs to our codebase” mentality.

For those of you that run things on autopilot, how do you keep code quality under control? And how do you handle refactoring? I am really curious, because one option now is also to just YOLO your LLMs to write code based on the maturity of the product. You can refactor an app or parts of it pretty fast again with LLMs. While tech debt accumulates faster, we also have the opportunity to rebuild faster.

Post reply on HN