Explainable Linear Programs(jeremykun.com)

80 points by matt_d 1 year ago | 23 comments

Prompting Is Programming(brethorsting.com)

5 points by aaronbrethorst 10 days ago | 0 comments

Blog post by Aaron Brethorst arguing that prompting an AI well is essentially programming, explaining why engineers get more value from LLMs than non-programmers.

What is “literate programming”? (2024)(pqnelson.github.io)

69 points by joecobb 11 months ago | 40 comments

Literate Programming: Articles(literateprogramming.com)

47 points by ddougj 3 years ago | 38 comments

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

104 points by alpaylan 9 months ago | 112 comments

Understanding DeepMind's sorting algorithm(justine.lol)

325 points by jart 3 years ago | 48 comments

Justine Tunney's blog post analyzing DeepMind's AlphaDev-discovered sorting kernels and how they apply to C library sorting code.

Prevent cognitive debt by manually retyping LLM-generated code(ankursethi.com)

502 points by mpweiher 3 months ago | 413 comments

A personal blog post arguing that developers should manually retype code generated by LLM coding assistants to keep understanding it and avoid cognitive debt.

Recent AI model progress feels mostly like bullshit(lesswrong.com)

474 points by paulpauper 1 year ago | 375 comments

LessWrong linkpost essay arguing recent AI model improvements feel overstated, based on experience building an AI codebase security-scanning startup.

Software may be getting worse, but it used to be horrible(bitlog.com)

282 points by jakevoytko 17 days ago | 309 comments

Blog post by Jake Voytko arguing that while software may be degrading today, software from the 1990s and 2000s was far buggier, citing crashes, lost work, and insecurity.

GPT-5.5 hallucinates 3x more than MIT-licensed GLM-5.2(arrowtsx.dev)

474 points by oshrimpton 4 months ago | 235 comments

Blog post arguing that bigger AI models are not better, comparing benchmark and hallucination rates of GLM-5.2, GPT-5.5, Opus 4.8, DeepSeek V4 and others.

Efficient algorithms, the Exponential Time Hypothesis, and losing the plot(argmin.net)

92 points by sebg 13 days ago | 59 comments

Ben Recht's arg min blog post on the history of defining algorithmic efficiency, critiquing OpenAI's claimed subexponential SAT algorithm as galactic.

AWS Bedrock to require sharing data with Anthropic for Mythos and future models

369 points by TomAnthony 5 months ago | 216 comments

Hacker News discussion of Anthropic requiring 30-day data retention for Mythos-class Claude models on AWS Bedrock, moving data outside AWS's security boundary.