Benchmarking Opus 5 on SlopCodeBench(github.com)

367 points by dhorthy 3 months ago | 100 comments

A markdown report in HumanLayer's GitHub repo benchmarking Anthropic's Claude Opus 5 model on the SlopCodeBench coding benchmark.

Getting AI to work in complex codebases(github.com)

442 points by dhorthy 1 year ago | 368 comments

A GitHub markdown document from HumanLayer describing advanced context engineering techniques for getting AI coding agents to work in complex codebases.

Why Software Factories Fail (or: harness engineering is not enough)(github.com)

351 points by dhorthy 3 months ago | 250 comments

A markdown essay in HumanLayer's GitHub repo arguing why AI-driven software factories fail and that harness engineering alone is not enough for coding agents.

Janas: Run LLMs larger than RAM by streaming experts from SSD(github.com)

207 points by khalidkhair 13 days ago | 77 comments

GitHub repository for Janas, a pure C inference engine that runs LLMs larger than RAM by streaming MoE experts from SSD, on CPU or GPU.

GPT-5-Codex-Mini – A more compact and cost-efficient version of GPT-5-Codex(github.com)

43 points by wahnfrieden 11 months ago | 29 comments

Codex Desktop continuously generates Crashpad pending dumps(github.com)

156 points by AIorNot 8 days ago | 72 comments

GitHub issue reporting that the OpenAI Codex Desktop app continuously writes Crashpad pending dump files, growing disk usage by over 5GB per day.

Show HN: VoidAccess – open-source dark web OSINT, from query to a graph(github.com)

100 points by coding-maniac 6 days ago | 11 comments

GitHub repository for VoidAccess, an open-source tool that automates dark web OSINT investigations, filtering noise and NSFW content and presenting results as a graph.

Show HN: Willow – Open-source privacy-focused voice assistant hardware(github.com)

492 points by kkielhofner 3 years ago | 121 comments

GitHub repository for Willow, an open-source, self-hosted voice assistant firmware for Espressif ESP BOX hardware that integrates with Home Assistant.

Show HN: Whiteboard (YC W26) – An open-source IDE for thoughtful software design(github.com)

395 points by sidharthkmenon 1 month ago | 129 comments

GitHub repository for Whiteboard, an MIT-licensed desktop IDE built on Code-OSS where developers and AI agents collaborate on software design via diagrams and diffs.

Show HN: PhreshOS – OS for Web Apps(github.com)

6 points by zouhirdx 24 days ago | 2 comments

GitHub repository for PhreshOS, an MIT-licensed open-source browser-based desktop system that hosts web apps as Node.js threads with built-in auth and real-time communication.

The GitHub wiki is an anti-pattern (2022)(michaelheap.com)

151 points by ibobev 1 month ago | 88 comments

A 2022 blog post by Michael Heap arguing that using the GitHub wiki for project documentation is an anti-pattern compared to a docs folder in the repo.

Show HN: We built an open source, zero webhooks payment processor(github.com)

375 points by agreeahmed 11 months ago | 209 comments

GitHub repository for Flowglad, an open source payments and billing infrastructure that integrates without webhooks and exposes real-time feature and usage-credit state.