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

Claude Code: An Agentic cleanroom analysis

southbridge-research.notion.site

Post reply on HN
Print view View original

51–58 of 58 posts

Re: Claude Code: An Agentic cleanroom analysis

#51

Post by insane_dreamer »

mindwok wrote: post #21
cedws wrote: post #15

Earlier quoted context omitted.

Could you briefly explain your workflow? I use Zed’s agent mode and I don’t really understand how people are doing it purely through the CLI. How do you get a decent workflow where you can approve individual hunks? Aren’t you missing out on LSP help doing it in the CLI?

Claude code has a VS Code plugin now that lets you view and approve diffs in the editor. Before it did that, I really don't understand how people got anything of substance done because it simply isn't reliable enough over large codebases.

Before that it showed you diffs in the console, which worked ok.

Claude Code now also has a PyCharm plugin (and probably other JetBrains IDEs) that also shows you diffs in the pycharm editor.

Re: Claude Code: An Agentic cleanroom analysis

#52

Post by scuff3d »

elliotec wrote: post #36
scuff3d wrote: post #30

Earlier quoted context omitted.

If this is what software engineering is going to become I'm finding a new job.

Better start now! It’s incredible and unbelievable how productive it is. In my opinion it still takes someone with a staff level of engineering experience to guide it through the hard stuff, but it does in a day with just me what multiple product teams would take months to do, and better. I’m building a non-trivial platform as a solo project/business and have been working on it since about January. I’ve gotten more d…

Hit me up when you release your product. I keep seeing stuff like this and never see any proof. Companies aren't releasing 10x the features/patches/bug fixes/products, open source isn't getting 10x the number of quality PRs, absolutely no real evidence that the massive productivity gains actually exist.

What I've seen is people feel more productive, until the reality of all the subtle problems start to set in. Even skilled engineers usually only end up with 10 or 20% productivity gains by the time they reduce its usecase to where it's actually not total dog shit, or by the time they go back around and fix all the problems.

The highest quality product I know of where the creator has talked about his use of AI is ghostty, and he's not claiming massive improvements, just that it's definitely helpful.

Re: Claude Code: An Agentic cleanroom analysis

#53

Post by SparkyMcUnicorn »

ramoz wrote: post #37
rane wrote: post #7

Earlier quoted context omitted.

Have you been able to interface Claude Code with Gemini 2.5 Pro? I'm finding that Gemini 2.5 Pro is still better at solving certain problems and architecture and it would be great to be able to consult directly in CC.

I do it indirectly. Gemini is my architecture goto. Claude Code for execution. It's just way more efficient to feed large portions of codebase at once to Gemini, pump out a plan and feed it to Claude Code. https://x.com/backnotprop/status/1929020702453100794

My most recent flow is very similar, but I use AiderDesk[0] instead of Prompt Tower for easier creation/editing of plan files.

AiderDesk lets you save snapshots of a point in time, so I create "presets" to restore sets of context files and/or conversation history (you can restore one or both), which is a really nice bonus. You can also add/remove context as needed without the manual copy/pasting work when I forget to include something or accidentally included too much. Its VS Code extension makes adding/removing files from context seamless.

[0] https://github.com/hotovo/aider-desk

Re: Claude Code: An Agentic cleanroom analysis

#54

Post by elliotec »

scuff3d wrote: post #52
elliotec wrote: post #36

Earlier quoted context omitted.

Better start now! It’s incredible and unbelievable how productive it is. In my opinion it still takes someone with a staff level of engineering experience to guide it through the hard stuff, but it does in a day with just me what multiple product teams would take months to do, and better. I’m building a non-trivial platform as a solo project/business and have been working on it since about January. I’ve gotten more d…

Hit me up when you release your product. I keep seeing stuff like this and never see any proof. Companies aren't releasing 10x the features/patches/bug fixes/products, open source isn't getting 10x the number of quality PRs, absolutely no real evidence that the massive productivity gains actually exist. What I've seen is people feel more productive, until the reality of all the subtle problems start to set in. Even s…

I’ll happily let you know when I release. Goal date for public beta is the 15th. I’d love eyes and feedback on it ASAP.

Hopefully it’s obvious that Claude will not have simply written the entire thing but you might get a sense of what it can do quickly as part of a whole - maybe similar to your last sentence but I suppose I am claiming massive improvements (in productivity, no warranty on quality yet).

Also keep in mind I’m entirely solo here. I fully agree with your points that the proof is in the pudding and obviously there’s nuance to all of it. But yeah, I’m not exaggerating with my commentary above.

Re: Claude Code: An Agentic cleanroom analysis

#55

Post by scuff3d »

elliotec wrote: post #54
scuff3d wrote: post #52

Earlier quoted context omitted.

Hit me up when you release your product. I keep seeing stuff like this and never see any proof. Companies aren't releasing 10x the features/patches/bug fixes/products, open source isn't getting 10x the number of quality PRs, absolutely no real evidence that the massive productivity gains actually exist. What I've seen is people feel more productive, until the reality of all the subtle problems start to set in. Even s…

I’ll happily let you know when I release. Goal date for public beta is the 15th. I’d love eyes and feedback on it ASAP. Hopefully it’s obvious that Claude will not have simply written the entire thing but you might get a sense of what it can do quickly as part of a whole - maybe similar to your last sentence but I suppose I am claiming massive improvements (in productivity, no warranty on quality yet). Also keep in m…

If you don't mind me asking a couple questions, what percentage of your code would you say is AI generated, meaning you promoted an AI and it went off and wrote code that you used (with or without modification)?

And how much time would you say you spend wrangling the AI, meaning either repromting or substantially editing what you get back?

Re: Claude Code: An Agentic cleanroom analysis

#56

Post by rane »

thegeomaster wrote: post #47
maleldil wrote: post #38

Earlier quoted context omitted.

Claude Code is "agentic". Aider isn't. It can plan, use external tools, run the compiler, tests, linters, etc. You can do some of it with Aider, too, but Claude is more independent. The downside is that it can get very expensive, very fast.

I've personally found that I reject around 80% of suggestions with "No, and tell Claude what to do differently". So it requires a lot of babysitting, and it usually means I cannot do another thing effectively while it's running. For this reason I've considered switching to something less agentic like Aider since it's more predictable. Curious to hear how others work around this.

Give better and more precise instructions.

Use the @ thing to prod it read some relevant files for context (kind of like with Aider)

Re: Claude Code: An Agentic cleanroom analysis

#57

Post by maleldil »

thegeomaster wrote: post #47
maleldil wrote: post #38

Earlier quoted context omitted.

Claude Code is "agentic". Aider isn't. It can plan, use external tools, run the compiler, tests, linters, etc. You can do some of it with Aider, too, but Claude is more independent. The downside is that it can get very expensive, very fast.

I've personally found that I reject around 80% of suggestions with "No, and tell Claude what to do differently". So it requires a lot of babysitting, and it usually means I cannot do another thing effectively while it's running. For this reason I've considered switching to something less agentic like Aider since it's more predictable. Curious to hear how others work around this.

My experience has been quite different. I often give it a large block of instructions and let it run autonomously for a while. When I come back it often did what I expected. It doesn't have good taste with respect to APIs, though, so sometimes I need a heavier hand on that.

I find it helps to have a CLAUDE.md file with instructions and thorough documentation. This is on a ~30k LOC Python codebase with type-checking and tests. YMMV with other languages.

Re: Claude Code: An Agentic cleanroom analysis

#58

Post by bananapub »

cedws wrote: post #15
triyambakam wrote: post #3

Claude Code with Sonnet 4 is so good I've stopped using Aider. This has been hugely productive. I've been able to write agents that Claude Code can spawn and call out to for other models, even.

Could you briefly explain your workflow? I use Zed’s agent mode and I don’t really understand how people are doing it purely through the CLI. How do you get a decent workflow where you can approve individual hunks? Aren’t you missing out on LSP help doing it in the CLI?

claude code churns away in a terminal, I have the git repository open in emacs with auto-revert-mode enabled so it reloads files if they're changed under it.

I view the files, and then review the changes in magit, and either approve some or all of them and commit them, or tell claude to do some thing else.

it works astonishingly well.

Post reply on HN