About the position
As the Engineering Manager for the GPU Kernel team, you’ll lead the team responsible for writing custom kernels and libraries which enable our transformer-based driving models to run efficiently on embedded GPUs and accelerators. This team works closely with ML engineers, software engineers and researchers to deploy end-to-end AI for autonomous vehicles at scale. This is an exciting opportunity to lead in several high impact, early stage projects at Wayve with the ultimate goal of enabling product deployments onto millions of customer vehicles around the world.
Responsibilities
- Lead a multi-disciplinary team of ML GPU kernel engineers to enable efficient ML deployments across millions of customer vehicles.
- Set key foundational strategy on deployment frameworks, compilers, toolchains and SoCs together with tech leads.
- Set clear objectives and priorities, and allocate resources efficiently.
- Develop new skills, especially within end-to-end ML and inference optimisation.
Requirements
- Proven experience as an Engineering Manager delivering complex engineering projects.
- Experience developing GPU kernels and/or ML compilers (e.g. CUDA, OpenCL, TensorRT, MLIR, TVM, etc).
- Experience optimising systems to meet strict utilisation and latency requirements.
- Excellent interpersonal and communication skills.
Nice-to-haves
- Experience with C++ and ML frameworks such as PyTorch.
- Experience with ML deployment pipelines.
- Experience with embedded SoCs used in automotive environments, e.g. Nvidia, Qualcomm, Renesas, etc.
