Notes from the
routing layer.
Engineering guides, agent infrastructure, and the operating principles behind Strix.
Which AI model should I use? A normal person's guide
A practical, mildly unhinged guide to choosing an AI model by task, speed, cost, and the consequences of being confidently wrong.
ECONOMICS02 / 2 MIN READThree ways Strix reduces coding-agent spend
A practical breakdown of payload filtering, capability routing, and measurable savings.
ENGINEERING03 / 1 MIN READWhat reliable agent infrastructure needs
Why model selection, provider health, and compatibility belong in an infrastructure layer.
GUIDE04 / 1 MIN READScale coding-agent access across a team
A clear operating model for shared usage, budgets, roles, and team-scoped API keys.
ROUTING05 / 1 MIN READModel routing is infrastructure, not a dropdown
Why coding agents need an intent and reliability layer between tools and model providers.
GUIDE06 / 1 MIN READConnect Cursor to Strix in five minutes
A concise walkthrough for using the Strix OpenAI-compatible endpoint and model IDs in Cursor.
ENGINEERING07 / 1 MIN READThe hidden coding-agent cost is often in the payload
How repeated context, tool schemas, and large tool results consume tokens before generation begins.