-
01
Open-Source Agent That Teaches Claude Code Your Architecture
AI made software cheap to build but not cheap to scale. Evolutionary architecture with domain-aware AI agents is the missing piece.
6min 1,567 w -
11
Claude Code Can Now Spawn Copies of Itself in Isolated VMs
Wiring up an MCP server so Claude Code can spawn isolated VMs, the SSE-to-Streamable-HTTP migration that broke everything, streaming architecture, and what productionising this would actually look like.
6min 1,446 w -
21
OpenHealth – Chat with Apple Health Data, Anywhere
I built OpenHealth because Claude's and ChatGPT's Apple Health connectors are US-only. Drop your export.zip, get seven markdown files any LLM can read — browser, CLI, or phone-to-desktop over WebRTC.
5min 1,155 w -
31
The Future of Security Is an Open-Source Model That Detects and Acts on Threats
Anthropic's Glasswing is for enterprises with six-figure security budgets. The rest of us need open-source security agents that learn our systems and act autonomously.
7min 1,695 w -
41
Claude Code Running Claude Code in 4-Second Disposable VMs
Anthropic has an internal platform for running Claude Code in full-permission isolation. I built my own version with Firecracker MicroVMs, Go, and a $400 Ryzen box.
7min 1,692 w -
51
What if your browser built the UI for you?
We're still shipping hand-crafted frontends while AI can generate entire interfaces. What if the browser itself generated the UI from an API manifest and your preferences?
5min 1,132 w -
61
NanoClaw's Architecture is a Masterclass in Doing Less
I pulled apart NanoClaw's 8,000-line codebase and found six architectural patterns that most SaaS teams should steal — credential proxies, container isolation as authorization, two-cursor message processing, and more.
6min 1,578 w -
71
Careers in the Self-Driving Codebase Era
The engineering career ladder was built for when code was expensive. Code is now cheap. What's valuable has fundamentally shifted.
6min 1,390 w -
81
What if a Worm Could Make AI Agents Smarter?
Introducing c302 — a research project that uses the C. elegans roundworm's neural dynamics as a behavioural control layer for an LLM coding agent.
5min 1,281 w -
91
Why LLMs waste half their brain
AI coding agents burn too many tool calls just figuring out where things are. I built codemap to fix that — cheap model indexes, cheap model selects, big model codes.
4min 956 w