Claude Design System Prompt(github.com)

103 points by handfuloflight 3 months ago | 28 comments

Riscrithm – An intuitive RISC-V assembler and optimizer coded in Go(github.com)

25 points by patrick-ghetea 5 months ago | 2 comments

Verifiable science on modified PCR machine(github.com)

45 points by kotaKat 1 year ago | 15 comments

DAPO: An Open-source RL System from ByteDance Seed and Tsinghua AIR(github.com)

13 points by the_arun 1 month ago | 2 comments

Show HN: Code Claude Code(github.com)

103 points by sean_ 1 year ago | 23 comments

A Simulation in C++ of Joseph Weizenbaum's 1966 Eliza(github.com)

28 points by m1guelpf 1 year ago | 5 comments

How we hacked McKinsey's AI platform(codewall.ai)

420 points by mycroft_4221 7 months ago | 168 comments

CodeWall blog post describing how its autonomous offensive agent found vulnerabilities in McKinsey's internal AI platform Lilli.

AI models can provably hide arbitrary capabilities(arxiv.org)

252 points by sbulaev 13 days ago | 118 comments

An arXiv paper showing AI models can embed encrypted neural network backdoors that provably hide arbitrary capabilities from elicitation-based safety evaluations.

Exploiting the most prominent AI agent benchmarks(rdi.berkeley.edu)

561 points by Anon84 6 months ago | 136 comments

UC Berkeley RDI blog post describing how researchers exploited major AI agent benchmarks like SWE-bench and Terminal-Bench to score highly without solving tasks, and proposed fixes.

What Security Teams Need to Know About Agents Using Abliterated Models(eve.security)

5 points by mooreds 17 hours ago | 0 comments

Eve Security blog post explaining abliterated open-weight models and how security teams can detect, govern, and block AI agents that use them.

What AI-Native Looks Like(fagnerbrack.com)

14 points by fagnerbrack 1 month ago | 7 comments

A blog post by Fagner Brack on Medium describing what an AI-native approach to software development with stable infrastructure looks like.

Local AI needs to be the norm(unix.foo)

1767 points by cylo 5 months ago | 699 comments

Blog post on unix.foo arguing that apps should run AI models locally rather than relying on API calls to OpenAI or Anthropic.