Senior Machine Learning Engineer – AI/ML Compiler
Core
Build and maintain machine learning compiler technologies that convert AI models (from PyTorch or ONNX) into efficient code for device chips (CPU, GPU, NPU).
Role type
Senior Machine Learning Engineer (AI/ML Compiler)
Builds
AI hub compiler, ONNX Runtime QNN, debugging tools for model compilation and validation
Domain
Mobile, edge, auto, and IoT hardware/software
Deliverable
production ML models
Required skills
ML compiler concepts (graph IRs, operator fusion, shape inference, lowering passes, backend partitioning), Python, C++, PyTorch model export, on-device deployment frameworks (LiteRT, ExecuTorch, ONNXRuntime), SoC-level constraints (memory bandwidth, compute precision, NPU/DSP execution), CI/CD pipelines for model compilation
Preferred skills
Industry experience in ML infrastructure, compiler engineering, or AI framework development, QAIRT/QNN familiarity, Git proficiency
Technologies
MLIR, ONNX, TVM, PyTorch, FX, ATen IR, LiteRT, ExecuTorch, ONNXRuntime, QAIRT, QNN
Responsibilities
Build & maintain ML compiler technologies to turn AI models into efficient device code; Contribute to AI hub compiler and ONNX Runtime QNN for graph optimization and partitioning; Build debugging tools to spot and fix failures, accuracy loss, or slowdowns; Solve open-ended problems independently while mentoring teammates; Explain complex compiler ideas to chip engineers, business partners, and outside developers
Seniority
Senior, hands-on IC with moderate organizational influence
