GitNation
Log in
Hamid Shoja

Hamid Shoja

I'm a full-stack engineer at Corsearch with deep proficiency in TypeScript and modern front-end frameworks. I work on building and integrating AI agentic applications in production, from custom AI-powered tools to MCP-based service integrations. I'm passionate about AI-assisted engineering and using it as a practical force multiplier to ship better software, faster.
Corsearch, Netherlandshshojahash01
Interests:
reactperformancenext.jsbest practicestypescriptjavascriptnode.jscode qualityweb developmentfrontend
Your Agent Docs Are Lying: Structuring and Testing Instruction Files for Coding Agents
AI Coding Summit BerlinAI Coding Summit Berlin
Upcoming
Your Agent Docs Are Lying: Structuring and Testing Instruction Files for Coding Agents
When we audited our AI coding-agent instructions (CLAUDE.md files, skills, agent definitions) against our actual monorepo, we found docs that were confidently wrong: styling guides producing CSS that never matched the build, test templates that crashed on render, "best practices" used by 1 of 50 real files. Every wrong doc cost a multi-thousand-token debug loop each time an agent followed it. This talk is the playbook that came out of fixing it: how to distribute knowledge across instruction surfaces based on their loading semantics, and how to test docs the way you test code.
artificial intelligence