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 an end-to-end learning approach. The ADAS engineering team develops perception and intelligence systems that power driver assistance in real-world driving scenarios.
In this role, you will train, debug, and improve computer vision and 3D perception models that enable vehicles to understand and navigate complex urban environments. You'll work across the full ML lifecycle—from data creation and pipeline development to model training, evaluation, and iteration based on clear performance signals.
Key responsibilities include:
- Developing core ADAS perception capabilities such as detection, classification, and instance segmentation for lanes, objects, traffic signs, and traffic lights
- Building end-to-end approaches that allow models to act on the world directly, reducing reliance on hand-coded intermediate steps
- Creating online perception systems that run efficiently in-car as part of the production stack
- Developing offline perception work using additional sensors (lidar, future frames, maps) and compute to improve data quality and coverage at scale
- Building data flywheels using offline models and 3D tracking/reconstruction to auto-generate labels and accelerate iteration
- Partnering with the team to prioritize work based on system performance gaps
- Building scalable data pipelines including auto-labelling and pseudo-labelling
Wayve operates a hybrid model across main hubs in London, Sunnyvale, Yokohama, Herzliya, Vancouver, and Leonberg, combining in-person collaboration with remote flexibility. The company emphasizes real ownership and pragmatic problem-solving under product constraints.
Requirements:
- Proven track record building and shipping CV-focused deep learning systems with strong applied ML engineering (not research-only)
- Experience with 3D perception concepts or pipelines such as LiDAR, multi-view geometry, tracking, or 3D reconstruction
- Ability to own work end-to-end, including evaluation and large-scale dataset generation
- Pragmatic problem-solving skills under real product constraints
- Passion for improving real-world driving performance through better perception
- Automotive domain experience is a bonus but not required if perception background is strong from other domains
- The company welcomes applications from candidates passionate about autonomy who may not meet every requirement
