Article
Lost in the Middle: Why Your LLM Ignores What You Give It
LLMs don't read long contexts well, not because of capacity limits, but by design. Two studies measure the gap. What it means for how you architect your systems.
azrod.me
Welcome. This blog is a space for technical reflection on AI agents, DevOps and modern software architecture. Articles are written in French and English.
Brève
pure.md — a markdown proxy for AI agents, with built-in MCP
pure.md converts any URL into LLM-ready markdown: bot detection bypass, JS rendering, global cache. Two MCP tools are all you need to plug it into Cursor, Windsurf, or Claude Desktop.
Topics
Latest articles
View all articles → Article
Token Economy: Three Tools Attacking the Problem Differently
RTK, DCP, Caveman — three approaches to reducing token consumption in AI agents. Not just about cost: primarily about reasoning quality.
Article
Agentic AI: what it actually is
Between devs convinced AI will automate everything and those who think it will never truly understand code, there's a more nuanced reality, and a more useful one.
Brève
pure.md — a markdown proxy for AI agents, with built-in MCP
pure.md converts any URL into LLM-ready markdown: bot detection bypass, JS rendering, global cache. Two MCP tools are all you need to plug it into Cursor, Windsurf, or Claude Desktop.
Brève
Muxy, a Mac terminal built around libghostty
A nice Mac find: Muxy builds on libghostty and leans into worktrees, vertical tabs, and splits.