WebPerf Platform Take-Home

Minimal platform for a small web performance monitoring company moving away from SSH-and-screen deployments.

Quick Start

Prerequisites: Docker Desktop or Docker Engine, Docker Compose, and make.

make up
make seed

Open:

• Dashboard: http://localhost:8080
• API docs: http://localhost:8000/docs
• Prometheus: http://localhost:9090
• Grafana: http://localhost:3000 (admin / admin)

What Runs

• api: FastAPI ingestion/config/dashboard API
• worker: Go background worker consuming queued events
• frontend: static dashboard served by Nginx
• redis: queue and small aggregate store
• prometheus: metrics scraping
• grafana: provisioned dashboard surface

Common Commands

make up        # build and run the full local platform
make down      # stop and remove containers
make logs      # follow logs for all services
make ps        # show running services
make seed      # send sample events through the API
make smoke     # quick health checks
make fail      # simulate a worker outage
make recover   # restart the worker

Demo Recording

Loom link: https://www.loom.com/share/2d267795f43c4d308565457c9544a00a

Suggested 5-minute flow:

1. Run make up, then make seed.
2. Show the dashboard and Grafana.
3. Run make fail and send more events with make seed.
4. Show queue depth rising in Grafana/Prometheus.
5. Run make recover and show the worker drain the queue.

Adding a Fourth Service

See docs/user-guide.md. In short: add a new directory under services/, give it a Dockerfile and health endpoint, add a services/<name>/compose.yml fragment, and add a Prometheus target file if it exposes metrics. make up discovers service compose fragments automatically, so no platform code has to change.

Docs

• Design doc
• Engineering user guide
• Failure runbook