AI glossary
Observability
Observability for AI features is recording each model call, its prompt, settings, tokens, time and outcome, so you can see what happened and why.
A trace follows one request through every step: retrieval, each model call, each tool call. You can see which step was slow, which was expensive and where an answer went wrong.
Useful fields are the model and prompt version, token counts, latency, cost, errors and a link to the user's feedback.
Logs of prompts can hold personal data. Redact before you store, keep them only as long as you need them, and limit who can read them.
Learn it hands-on
These free lessons teach it with commands to run in a lab and checked tasks. Read them whole with no account.