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

Please stop flooding our projects with AI slop to furnish your CV

neilalexander.dev

Post reply on HN
Print view View original

51–60 of 162 posts

Re: Please stop flooding our projects with AI slop to furnish your CV

#51

Post by jspdown »

nubinetwork wrote: post #36

This is nothing new, people have been making spellcheck commits all over github for years... and that was before ai slop even existed.

I help maintain a big project as well and I can totally relates what the author is describing.

The amount of security advisories and PRs opened is getting out of hands. This is no longer just spell check PRs that you can merge in less than 1 minute of review. Often that's going to be supposed perf improvement, attempts at fixing things that are not broken, or submitting completely new features.

These relatively low effort PRs takes a lot of effort to review or even just to filter out.

Re: Please stop flooding our projects with AI slop to furnish your CV

#53

Post by n_plus_1_acc »

bhaak wrote: post #48
J253 wrote: post #41

I also maintain a popular open source repo and have seen similar. If I receive a low effort single-pass obviously-Clauded PR (many people (agents) don’t even bother using the AGENTS.md file I provide right in the repo?!) with no associated issue I also have no qualms closing. I do generally leave a short note about why I closed (volume of these is around 5/week) showing how far off they (Claude) was and encourage ope…

> If I receive a low effort single-pass obviously-Clauded PR (many people (agents) don’t even bother using the AGENTS.md file I provide right in the repo?!) Claude doesn't read AGENTS.md automatically, it only looks for CLAUDE.md.

Very intelligent

Re: Please stop flooding our projects with AI slop to furnish your CV

#54

Post by fhd2 »

oleg_antonyan wrote: post #43

Do recruiters actually care about your open source contributions? Especially now when only LLMs read CVs and match them against strict criteria, I doubt there are many companies that actually care about your work outside of work, in fact they might care in opposite direction - thinking that you'll be distracted from work

Just speculating (I use zero automation when recruiting), but automated candidate ranking systems could well still look for signals that were valuable years ago, but no longer are. It's not a given that people using this stuff even understand the criteria under the hood.

Same story as SEO, I presume. Google surely understood their ranking criteria deeply, yet couldn't exactly win that fight.

Re: Please stop flooding our projects with AI slop to furnish your CV

#55

Post by alkonaut »

Ironically this seems like a perfect use for AI from the maintaner side. "Find low-effort/AI-like PR's without associated issues from new contributors. Discuss with the contributors in the PR about how contributions should be made, then reject their PR's. If any bad behavior is detected in this interaction, block the accounts for 24 hours and add them to a list of accounts to review for permanent banning from the project".

Automated PR's can have automated responses. You make a human effort, you get a human response.

Re: Please stop flooding our projects with AI slop to furnish your CV

#58

Post by J253 »

bhaak wrote: post #48
J253 wrote: post #41

I also maintain a popular open source repo and have seen similar. If I receive a low effort single-pass obviously-Clauded PR (many people (agents) don’t even bother using the AGENTS.md file I provide right in the repo?!) with no associated issue I also have no qualms closing. I do generally leave a short note about why I closed (volume of these is around 5/week) showing how far off they (Claude) was and encourage ope…

> If I receive a low effort single-pass obviously-Clauded PR (many people (agents) don’t even bother using the AGENTS.md file I provide right in the repo?!) Claude doesn't read AGENTS.md automatically, it only looks for CLAUDE.md.

True, but I do have a claude.md as well. Point is, best practice agentic development doesn’t seem to be on display.

Re: Please stop flooding our projects with AI slop to furnish your CV

#59

Post by geraneum »

arjie wrote: post #18

This makes sense. Apart from the original thing, I no longer upstream anything. It just takes comparatively more effort than maintaining a private fork with my own idiosyncratic fix. This must mean that even small fixes that non-contributors would make in the past are just happening off the books, so to speak. e.g. I upstreamed a tiny fix to Thrift codegen because some function could be much faster. These days I woul…

Discussing the issues, achieving consensus, making sure the fix fits the existing patterns and actually warrants the change, etc. These are and have always been a challenging part of contributing to open source. Typing code in your private fork that satisfies your “idiosyncratic fix” has always existed and been always the easier aspect of such changes. Chances are many of these changes would never be accepted into the repository anyway in the past.

Re: Please stop flooding our projects with AI slop to furnish your CV

#60

Post by jspdown »

ReptileMan wrote: post #34

Create AI to review the pull requests and auto ban people for low SNR.

AI code review is nice to have an additional pair of eyes but it doesn't substitute to the maintainer eyes. A big part of a review is deciding if you want that change merged in. Not because of the immediate code but what it means to the project to bring this in.

AI code review, to my eyes, just reduce the number of bugs, they do not shorten the review time. Otherwise, it means you are compromising on the direction of the project.

Post reply on HN