Your mission
To build and lead Track Titan's ML capability at a pivotal moment for our AI direction. You will take our unique dataset – millions of laps of racing telemetry across thousands of car/track combinations – and turn it into the foundation of a genuinely new kind of AI coaching platform. You'll be hands-on technically from day one, while growing and guiding a small team that will define how we do ML at Track Titan.
What you will do:
- Build the telemetry understanding layer. Design and train the models that learn meaningful representations from racing telemetry – self-supervised encoders, sequence models, embedding systems. This is the technical foundation everything else builds on, and you'll own it end-to-end: from problem formulation through architecture decisions to production deployment.
- Evolve our coaching intelligence. We have an existing coaching system and a massive dataset – your job is to unlock the next level. You'll introduce new approaches to problems like mistake identification and time-loss attribution that scale with our data, moving us from domain-engineered solutions toward learned models that generalise across cars, tracks, and driving styles.
- Set the ML technical direction. Evaluate approaches, define the model and data strategy, and make the architectural decisions that will shape our ML stack for years. Establish the practices that make ML work reliable – experiment tracking, evaluation harnesses, reproducibility. You'll work closely with the CTPO (who has deep AI/ML experience) to pressure-test ideas and align ML work with product and business priorities. You'll have strong opinions, the technical depth to back them up, and the humility to change your mind when the data says otherwise.
- Lead and grow the ML team. You'll lead a small, growing ML team and be responsible for raising the bar: code quality, experimentation rigour, review culture, and technical development. As we scale, you'll shape what the ML function looks like and play a key role in hiring the people who join it.
- Ship models that reach users. Research taste matters, but so does pragmatism. You'll balance long-term foundational work with near-term product impact, finding the stepping stones where better models directly improve insights and coaching quality for hundreds of thousands of drivers. You'll define clear contracts with the product engineering team and own ML systems through to production.