// the find
slowlydev/f1-dash
A real-time F1 dashboard
Real-time F1 telemetry dashboard that pulls live session data and displays leaderboard, tire strategy, mini sectors, gaps, and car metrics. The backend is a Rust service that connects to the F1 SignalR feed and rebroadcasts over WebSockets; the frontend is Next.js. For F1 fans who want more than the official app gives them during a race weekend.
The Rust backend is a genuinely good architectural call — parsing and rebroadcasting high-frequency telemetry in TypeScript would be painful, and the Rust SignalR client they wrote themselves (the `signalr` crate in the repo) handles reconnection cleanly. The state merging logic (`lib/merge.ts`, `useDataStore`) handles partial F1 state updates correctly, which is non-trivial given the F1 feed sends diffs not snapshots. The replay/simulator tool is a real development win — you can feed recorded sessions back through the stack without waiting for a live race. The component split between leaderboard, driver detail, and track map is sensible and the hooks (`useDataEngine`, `useBuffer`) hide the WebSocket complexity from UI components.
The README is nearly empty — no explanation of what data the F1 feed actually provides, no architecture diagram, and SETUP.md carries all the weight. Self-hosting this requires running three separate services (realtime, api, dashboard) and the Kubernetes manifests in `.k8s/` suggest that's the intended production path, but there's no guidance for anyone who just wants docker-compose. The F1 SignalR endpoint is undocumented and unofficial, so this breaks silently when F1 changes their stream format mid-season — there's no versioning or schema validation layer to catch that early. Zustand stores (`useDataStore`) hold the entire session state in one flat object; as the data model grows this is going to get unwieldy and cause unnecessary re-renders across unrelated components.