News for hackers. Stuff that matters.
Re:Claude Sonnet 4.5 by sigmoid10 on Monday September 29, 2025 @02:27PM (#45417094)
This is obviously much more than just taking an LLM an letting it run for 30 hours. You have to build a whole environment together with external tool integration and context management and then tune the prompts and perhaps even set up a multi-agent system. I believe that if someone puts a ton of work into this you can have an LLM run for that long and still produce sellable outputs, but let's not pretend like this is something that average devs can do by buying some API tokens and kicking off a frontier model.
Re:Claude Sonnet 4.5 by Philpax on Monday September 29, 2025 @02:40PM (#45417244)
Well, yes, that's Claude Code. And OpenAI Codex. And Google Gemini CLI.

Your average dev can just use those.

Re:Claude Sonnet 4.5 by ewoodrich on Monday September 29, 2025 @05:39PM (#45419098)
But then that goes back to the original question, considering my own experiences observing the amount of damage CC or Codex can do in a working code base with a couple tiny initial mistakes or confusion about intent while being left unattended for ten minutes, let alone 30 hours....
Re:Claude Sonnet 4.5 by janee on Tuesday September 30, 2025 @12:30AM (#45421903)
Yes but you need to setup quite a bit of tooling to provide feedback loops.

It's one thing to get an llm to do something unattended for long durations, it's a other to give it the means of verification.

For example I'm busy upgrading a 500k LoC rails 1 codebase to rails 8 and built several DSLs that give it proper authorised sessions in a headless browser with basic html parsing tooling so it can "see" what affect it's fixes have. Then you somehow need to also give it a reliable way to keep track of the past and it's own learnings, which sound simple but I have yet to see any tool or model solve it on this scale...will give sonnet 4.5 a try this weekend, but yeah none of the models I tried are able to produce meaningful results over long periods on this upgrade task without good tooling and strong feedback loops

Btw I have upgraded the app and taking it to alpha testing now so it is possible

Re:Claude Sonnet 4.5 by grncdr on Tuesday September 30, 2025 @02:52AM (#45422662)
> rails 1 codebase to rails 8

A bit off topic, but Rails *1* ? I hope this was an internal app and not on the public internet somewhere …

Re:Claude Sonnet 4.5 by janee on Tuesday September 30, 2025 @05:04AM (#45423444)
haha no it's an old (15years old) abandoned enterprise app running on-prem that hasn't seen updates in more than a decade.
Re:Claude Sonnet 4.5 by sarchertech on Tuesday September 30, 2025 @07:22AM (#45424181)
Wow Rails 3 came out 15 years ago, so that thing started life out of date.
Re:Claude Sonnet 4.5 by lsaferite on Wednesday October 01, 2025 @09:21AM (#45437401)
> enterprise app

> started life out of date

That tracks my experiences.

Re:Claude Sonnet 4.5 by majortennis on Wednesday October 01, 2025 @10:28AM (#45438132)
I've tried asking it to log every request and response to a project_log.md but it routinely ignores that.

I've also tried using playwright for testing in a headless browser and taking screenshots for a blog that can effectively act as a log , it just seems like too tall an order for it.

It sounds like you're streets ahead of where I am could you give me some pointers on getting started with a feed back loop please

Re:Claude Sonnet 4.5 by sigmoid10 on Thursday October 02, 2025 @03:39AM (#45447181)
If you had used any of those, you'd know they clearly don't work well enough for such long tasks. We're not yet at the point where we have general purpose fire-and-forget frameworks. But there have been a few research examples from constrained environments with a complex custom setup.
Re:Claude Sonnet 4.5 by ChadMoran on Monday September 29, 2025 @09:16PM (#45420902)
Claude Code with a good prompt can run for hours.