Machine Learning & Deployment
CompleteWhen should a driver pit? Predicting optimal pit-stop timing from race conditions.
A dual-model XGBoost system predicting optimal pit-stop lap from 2,600+ historical stops, served through a FastAPI backend with Docker and CI/CD.
Pit timing balances tire degradation, track position, and race conditions — get it wrong and you lose places; get it right and you win them. This project predicts the optimal pit lap from the current race state.
It doubles as a study in model selection: does more data always help? The answer here is instructive — a focused single-season model beat a broader multi-season one.
A personal ML project, deployed as a working API and dashboard.
| Table | Description | Size |
|---|---|---|
| stint | Which tire stint the driver is on — the top predictor (38.7%). | feature |
| compound / tyre_life | Tire type and age. | feature |
| safety_car / position | Race conditions and track position. | feature |
The single-season model (R² 0.851) outperformed the multi-season model (R² 0.772) by roughly 10% because a single regulation era is more homogeneous.
Why it matters. It is a useful counter-example to “add more data”: matching the training window to the decision context beat simply using more history.
Caveat. The single-season model is narrower — it assumes the current regulations hold.
Feature importance is led by stint number (38.7%) and tire compound (18.6%), with safety-car status a distant third (9.3%).
Why it matters. These are exactly the levers a race engineer reasons about, so the model’s logic is legible rather than a black box.
Single-season (R² 0.851) versus multi-season (R² 0.772) quantified the cost of regulation heterogeneity directly.
Feature importance is led by stint and tire compound — matching how race engineers actually reason about pit timing.
Served through a FastAPI backend with sub-50ms average response times, containerized with Docker and deployed with CI/CD.
The single-season model maximizes accuracy under current regulations; the multi-season model trades accuracy for robustness across rule changes. Offering both makes the accuracy/robustness trade-off explicit.
FastAPI backend (Pydantic validation, OpenAPI docs), Docker image (~180MB), deployed on Railway with GitHub Actions CI/CD.