RaceNebula

Inspiration

We’re obsessed with the split-second strategy calls that decide F1 races. Teams rely on massive data streams, but fans and developers rarely see that intelligence. We wanted to recreate that thrill — bringing live, data-driven strategy prediction to everyone.

What it does

RaceNebula is an AI-powered F1 strategy platform that ingests live telemetry to predict pit stops, tire degradation, overtakes, and race outcomes — all visualized in a cosmic, interactive dashboard. It helps strategists and fans anticipate decisions before they happen.

How we built it

We parsed second-by-second telemetry from F1 sessions using Python (Pandas, NumPy) and AWS for data handling. Our machine learning models predict race-critical events like pit stop windows and tire wear. The front end, built with React + Tailwind, transforms complex data into a sleek, space-themed experience.

Challenges we ran into

Handling multi-driver time-series data was tough — millions of rows with overlapping laps. Building efficient parsing, smoothing noisy telemetry, and aligning asynchronous events pushed our data engineering and ML pipelines to the limit.

Accomplishments that we're proud of

Built a complete end-to-end F1 race intelligence system in under 48 hours

Designed a real-time dashboard with race playback and predictive overlays

Created high-accuracy tire degradation models and pit stop predictors

Achieved a beautiful cosmic UX that makes data analysis feel like space exploration

What we learned

We learned how to manage huge racing datasets, design real-time ML inference pipelines, and balance performance with visualization. Most importantly, we realized how data can make sports strategy feel alive.

What's next for RaceNebula

We’re expanding into live streaming APIs, reinforcement learning for race simulation, and user-driven custom strategy testing. Eventually, we want RaceNebula to become the go-to open platform for motorsport analytics and AI racing strategy.