About the job
As a Staff Machine Learning Engineer on Wayve's Measurement team in AI Evaluation, based in our Sunnyvale office, you will build the computer vision and scene understanding models Wayve uses to measure the performance of the Wayve Driver offline. You will adapt technology from our on-vehicle models and Wayve Foundation Models into offline models that understand coverage, mine rare events, and assess driving behaviour, and you will drive their accuracy and generalisation across vehicles, markets, and conditions.
Responsibilities
Develop the models - build, train, and fine-tune the scene understanding models at the centre of Wayve's offline measurement, adapting on-vehicle architectures and Wayve Foundation Models for offline use.
Drive accuracy and generalisation - improve model performance across vehicle platforms, geographies, and driving conditions; diagnose failure modes and close the loop on blind spots.
Exploit the offline environment - use the advantages the vehicle does not have: higher compute budgets, larger model capacity, bidirectional temporal context, and multi-task or joint representation learning.
Measure what you build - benchmark your models, set quality bars, and use metrics and error analysis to steer the next iteration; treat measurement as the feedback that drives the modelling.
Make the evidence credible - ensure benchmarked results are statistically defensible and fit to feed validation pipelines at scale and our broader safety cases, across the product portfolio.
Align priorities and mentor - work day-to-day with on-vehicle modelling, evaluation, data curation, and simulation teams across sites; raise the bar on engineering and modelling practice; mentor others on the team; keep sight of division and company priorities and how Measurement work enables them.
Qualifications
Minimum
5+ years in ML engineering, including training and shipping deep learning models in production, with pathfinding in ambiguous modelling problems from scoping through to a direction others build on.
Hands-on experience training modern computer vision models, including transformer-based and multimodal or VLM architectures for detection, segmentation, classification, or scene understanding, on camera and/or lidar sensor data.
Experience adapting or fine-tuning large pretrained or foundation models, and training shared representations across multiple tasks or objectives (multi-stage or joint training), including real trade-offs across data and losses.
Proficient in Python and ML frameworks (esp. PyTorch), with a solid foundation in software engineering practices and comfort with large-scale training.
Staff-level technical leadership: research-literate and pragmatic, setting direction, raising the bar, and leading cross-functional work without formal line management.
Able to measure your own models: comfortable defining and reading the metrics that show whether a model is genuinely improving.
Preferred
Experience in 3D scene understanding and representation learning for geometric and semantic perception, including large-scale semantic enrichment of driving scenes.
Experience with offboard or offline modelling: auto-labelling, model distillation, temporal or world models, or other ways of exploiting compute that on-vehicle systems cannot.
Prior experience in autonomous vehicles or robotics with hands-on deployment and closed-loop validation on physical systems.
Experience with fleet-scale data and large-scale distributed training infrastructure.