tool use

Language models predict tokens rather than calculate. Let the model extract numbers and let code do the maths, then check the result against the printed total.

Agent evals are where answers that worked for chatbots stop working. A correct final booking can hide a path that broke policy, spent money twice or took nine minutes.

Each tool has a different cost driver: search licences per query, retrieved content as tokens, containers by time. Price each on its driver, decide whether retrieved text is billed, and bundle cheap tools with expensive ones.

Agent prompts tempt candidates into ambitious builds that collapse. The strong answer builds the least autonomous version that demonstrates the idea, with mocked tools, clear limits and tests of the hard paths.