We gave terabytes of CI logs to an LLM(mendral.com)

198 points by shad42 8 months ago | 98 comments

Use Prolog to improve LLM's reasoning(shchegrikovich.substack.com)

343 points by shchegrikovich 2 years ago | 122 comments

Substack post arguing that having LLMs generate Prolog code can improve their reasoning, given the limits of autoregressive transformer generation.

LLMs can be exhausting(tomjohnell.com)

256 points by tjohnell 7 months ago | 171 comments

LLMs could be, but shouldn't be compilers(alperenkeles.com)

104 points by alpaylan 8 months ago | 112 comments

Anti-AI Hype LLM Reading List(gist.github.com)

152 points by atg_abhishek 3 years ago | 38 comments

Write Dumb Code (2018)(matthewrocklin.com)

99 points by dvcoolarun 2 years ago | 60 comments

If you are useful, it doesn't mean you are valued(betterthanrandom.substack.com)

748 points by weltview 1 year ago | 334 comments

A Better than Random Substack essay arguing that being useful at work is different from being valued, and that the distinction can be subtle.

Every SaaS business will become a harness around a model(blog.sshh.io)

117 points by iacguy 23 days ago | 85 comments

Substack essay by Shrivu Shankar arguing that every SaaS business will become a harness wrapping AI models with tooling and state.

There is no epidemic of loneliness, but there is an epidemic of scurvy(experimental-history.com)

59 points by luu 1 month ago | 31 comments

An Experimental History Substack essay by Adam Mastroianni arguing there is no loneliness epidemic but there is an epidemic of scurvy.

Loyal workers are selectively and ironically targeted for exploitation(sciencedirect.com)

341 points by ck45 3 years ago | 219 comments

Abstract page for a Journal of Experimental Social Psychology paper finding that loyal workers are selectively targeted for exploitation by managers.

Why I'm still bearish on LLMs after Navier-Stokes(dank.systems)

426 points by jaykru 1 month ago | 552 comments

Personal blog essay by Jay Kruer arguing LLMs remain overhyped despite headline results like Navier-Stokes, citing poor generalization, reward hacking, and specification costs.

TODOs aren't for doing(sophiebits.com)

380 points by todsacerdoti 1 year ago | 218 comments

A blog post by Sophie Alpert arguing that TODO comments are valuable as documentation of known issues even if never acted on, and shouldn't be auto-deleted or tracked.