Open-source metrics and time-series database.
The metrics backbone of cloud-native systems. Prometheus scrapes your services on an interval, stores time-series data, and exposes a powerful query language (PromQL). In the bootcamp we instrument the agent capstones with custom Prometheus metrics so students see what production observability actually looks like.
Request rate, p95 latency, token usage by model, cost per request, agent loop depth, tool-call success rate, retrieval recall, error rate by category. The bootcamp's observability module covers each one.
Updates from the AI world — what shipped, what we’re using in production, and what’s worth your attention. Two emails a month, no spam.
