signal over noise
A newsletter, blog, and course that keeps what actually matters in AI, open source, and dev tooling.
Everything here lives under a handful of durable pillars, so it stays useful long after you read it.
Notable open-source projects and releases in AI and dev tooling: what they do and why they matter.
Practical AI: agents, coding assistants, RAG, and evals. Applied, not hype.
Editors, CLIs, infra, and the craft of building. The terminal-and-.dev ethos.
The good stuff, explained: clear breakdowns of a concept, protocol, or release.
A curated digest: the best links, releases, and reads worth keeping.
[ + ]Every pillar, every post. The full cache.
A teardown of the five places retrieval quietly breaks between demo and production — chunking, stale indexes, near-duplicate context, recall vs. answer quality, and 'fine on 10 questions' — each with the fix.
How to pick a local LLM runner by the job, not the benchmark: Ollama, llama.cpp, and the alternatives across local hacking, embedding in an app, CPU-only boxes, and GPU servers.
TDD's failing-test-first loop, applied to AI: write the eval before the prompt, let the score say when an LLM feature actually works, and stop shipping changes you can't measure.
in the works
A developer's guide to shipping with LLMs & agents. Go from “I read about it” to “I shipped it.” A practical, open-source-first path to building real software with AI. No hype, lots of code.