Engineering notes on the economics of AI coding agents.https://decost.dev/https://decost.dev/blog/what-a-coding-agent-session-costs/https://decost.dev/blog/what-a-coding-agent-session-costs/A single prompt to Claude Code or Codex becomes many billed API calls, each re-sending the whole conversation. Here is the anatomy of that bill: the four token classes, their price multipliers, and the six things that make one session cost ten times another.Mon, 31 Aug 2026 00:00:00 GMThttps://decost.dev/blog/what-your-developers-see-is-not-what-you-pay/https://decost.dev/blog/what-your-developers-see-is-not-what-you-pay/Claude Code computes its cost figure locally, at list price, from one machine's history. Four documented ways that diverges from your invoice — and what a savings claim needs to survive a finance review.Mon, 31 Aug 2026 00:00:00 GMThttps://decost.dev/blog/gateway-build-vs-buy/https://decost.dev/blog/gateway-build-vs-buy/Most teams routing Claude Code and Codex through a gateway are choosing between LiteLLM, Bifrost, a cloud vendor's proxy and their own. The decision turns on five properties, and one of them — byte fidelity — quietly decides your cache bill.Sun, 30 Aug 2026 00:00:00 GMThttps://decost.dev/blog/uber-software-factory-playbook/https://decost.dev/blog/uber-software-factory-playbook/Uber grew AI coding usage 7× while unit costs fell by half. Their levers are public, mechanical and mostly configuration. Here is each one, what it is worth, what it costs to adopt, and the case against adopting it blindly.Sun, 30 Aug 2026 00:00:00 GMT