ML Compiler Engineer

Develop and enhance compiler backends to solve resource allocation challenges for low-level instructions in spatial architectures.

Responsibilities

• Develop model partitioning approaches including pipelined, tensor, model, and data parallelism.
• Perform hardware resource allocation, tiling, memory management, scheduling, and optimization for latency, bandwidth, and throughput.
• Map low-level instructions to hardware resources to ensure optimal performance.
• Collaborate with compiler developers to deliver production-grade solutions.

Required Skills

• 7+ years of industry experience with a Bachelor’s degree in Computer Science, or 5+ years with an MSCS.
• Proficiency in Modern C++ with the ability to deliver production-quality code.
• Experience with compiler infrastructures such as LLVM or MLIR.
• Experience with machine learning frameworks including ONNX, TensorFlow, or PyTorch.
• Experience in production compiler development.

Preferred Skills

• Algorithm design experience from high-level conceptual design to implementation.
• Experience with Open Source ML projects such as Torch-MLIR, ONNX-MLIR, Caffe, or TVM.