SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Wayve is building an AI platform for autonomous driving using end-to-end learning from real-world experience. The Driving Product team develops the AI driver's core behaviors, taking models from data through training, evaluation, and on-road validation, shipping them into production vehicles on monthly timelines. This is a small, high-ownership team at a point of rapid growth, blending ML research with shipping-focused engineering.
In this role, you will:
- Develop AI driver model architecture and training algorithms to introduce and enhance driving behaviors across Level 2, 3, and 4 autonomy
- Own key parts of the training model lifecycle, including evaluation strategy, success metrics, and iteration planning
- Mine, bucket, and curate real-world and synthetic data to teach specific driving behaviors, implementing data schemes to support training
- Run and analyze on-road and offline experiments, translate results into clear next steps, and drive improvements through repeated training cycles
- Work on models that predict vehicle trajectories and interact directly with controllers and actuators, closing the loop between prediction and real-world vehicle performance
- Build production-ready ML systems with strong software engineering, quality, and reliability practices
- Develop methods spanning reinforcement learning, behavioral cloning, and distillation, working with multi-sensor inputs (camera, radar, lidar)
- Establish on-road and offline metrics that diagnose issues and drive data-led model iterations
Wayve operates a hybrid working model across main hubs in London, Sunnyvale, Yokohama, Herzliya, Vancouver, and Leonberg, combining in-person collaboration with remote flexibility.
REQUIREMENTS:
- Proven track record training deep learning models with clear end-to-end ownership: data, training, evaluation, and iteration
- Experience taking ML models into production, working through real-world constraints, quality, and safety requirements
- Strong Python skills and hands-on experience with modern ML frameworks
- Comfortable shipping production-facing ML, not just research or proof-of-concept work
- Background in either end-to-end autonomous driving/embodied AI, or large-scale product ML systems (recommendations, ranking)
- Reinforcement learning or behavioral cloning experience, especially where it materially improved real-world performance
- Familiarity with model compression and deploying ML to embedded systems (nice to have)
The company notes that not meeting every requirement is acceptable if you are passionate about autonomy and keen to learn.
