Engineering
Route every Claude Code message to the right model with Jev
jev-router asks Jev, a System One model, which Claude model each Claude Code message needs, and routes it there for a fraction of a cent.
Engin Diri
9 min readEngin Diri
9 min read
Best Practices
10 tips to improve your coding agent game
Ten coding agent habits checked against the research behind them, from AGENTS.md drift and /compact to model switching, subagents, and review.
Engin Diri
14 min read
Perspectives
Superpowers, GSD, and GSTACK: Picking the Right Framework for Your Coding Agent
Superpowers, GSD, and GSTACK compared and updated: what each does, whether they work with Codex, and the alternatives worth knowing about.
Engin Diri
14 min read
Best Practices
The Claude Skills I Actually Use for DevOps
Skills teach AI agents how to work like experienced practitioners. In this post, we share several skills that can improve how you build cloud infrastructure.
Engin Diri
16 min readProduct
Pulumi Neo Team
Pulumi Agent Skills: Best practices and more for AI coding assistants
Introducing packaged Pulumi expertise that works across Claude Code, Cursor, GitHub Copilot, and other AI coding assistants.
Pulumi Neo Team
TutorialsEngin Diri
Self-Verifying AI Agents: Vercel's Agent-Browser in the Ralph Wiggum Loop
Vercel's agent-browser uses 82% less context than Playwright MCP. I tested both on the same six browser tests to see how they compare.
Engin Diri
TutorialsEngin Diri
How Ralph Wiggum Built a Serverless SaaS with Pulumi
See how an AI agent using the Ralph Wiggum loop built a full AWS serverless SaaS with Pulumi—unsupervised. Frontend, backend, tests, and CloudFront included.
Engin Diri