Senior ML Compiler Engineer

What you'll do

• Build and evolve the model compilation toolchain for deploying perception, prediction, and planning models to autonomous vehicles.
• Architect new compiler passes and analyses to improve build times, memory footprint, and runtime latency.
• Develop a pipeline to convert high-level model graphs into optimized inference artifacts for specific hardware accelerators.
• Collaborate with hardware and runtime teams to co-design interfaces and expose optimal abstractions for peak performance.
• Establish standards and best practices for model export, validation, and debugging for internal machine learning teams.
• Build robust tooling to validate numerical correctness and detect performance regressions in the compilation pipeline.
• Provide clear diagnostics and tools to help model authors identify issues during the deployment process.

What we're looking for

• Must have at least 3 years of experience in the field of compilers.
• Experience with ML frameworks such as PyTorch, TensorFlow, or JAX is required.
• Familiarity with software stacks including ONNX, MLIR, XLA, TVM, and TensorRT is required.
• Expertise in writing production-quality Python and C++ code is required.
• Must have expertise in the software development life cycle, including coding, debugging, optimization, testing, and integration.
• A BS or higher degree in CS, CE, EE, or an equivalent field is required.