Optimize Waabi’s distributed machine learning training with profiling, CUDA, and inference technologies across research and production.
This Senior or Staff engineering role focuses on creating stable, productive training frameworks for Waabi’s autonomous transportation technology. You will investigate CPU and GPU performance, assess quantization-aware training, and explore compilation and deployment techniques. Collaboration with researchers and machine learning engineers will guide optimal resource usage and practical adoption through tooling and dashboards.
Key Responsibilities: Drive distributed training toward greater stability and efficiency
Comprehensively profile runtime and memory performance
Identify technologies for training and inference frameworks
Design new CUDA kernels when appropriate
Improve dashboards that encourage broad tool adoption
Requirements: Computer Science, Robotics, or similar technical education
Minimum four years industry experience with bachelor's degree
Strong coding skills across Python, C++, or Rust
Deep learning framework experience with PyTorch or Jax
Profiling experience with PyTorch Profiler or NVIDIA Nsight
Bring distributed training, performance profiling, and framework optimization experience to Waabi’s growing engineering team.