12 hours ago
Responsibilities
- Design and build scalable, self-service AI performance tools across models, hardware targets, and development workflows.
- Define standards for measuring and predicting AI performance across the organization.
- Model theoretical platform peaks, compare them with achieved performance, and identify efficiency losses at the layer and operator levels.
- Predict model latency, memory, utilization, and compute cost before running expensive changes.
- Own monitoring and regression alerting for model builds and training runs.
- Collaborate with training and runtime engineers to establish performance targets and use data to influence priorities.
Requirements
- Deep hands-on experience in performance engineering for complex systems, including profiling, roofline analysis, latency and throughput optimization, and workload bottleneck analysis.
- Experience owning a tool or service end to end, from design and delivery through adoption by other teams.
- Strong Python skills and experience profiling and instrumenting large production codebases.
- Hands-on experience developing deep learning models with PyTorch.
- Strong data analysis skills and the ability to derive defensible conclusions from noisy measurements.
- Ability to turn ambiguous performance questions into measurable problems and prioritize impactful work.
- Clear quantitative communication skills for influencing other teams’ priorities.
Benefits
- Full-time role based in-office with a hybrid working policy combining office and workshop time with work from home.
- Inclusive interview experience with accommodations or adjustments available upon request.
- Inclusive and respectful workplace committed to diversity and equal treatment.
Categories
About Wayve
Wayve builds end-to-end autonomous driving software—the vehicle-agnostic Wayve AI Driver—that runs on onboard compute and native sensors, licensed to automakers and fleet operators. Its platform spans ADAS and higher autonomy (L2+/L3 to robotaxi) and is designed to generalize across vehicle types and geographies. Founded in 2017 and headquartered in London, it tests its models across Europe, North America, and Japan, with a U.S. base in Sunnyvale, CA.