Staff Machine Learning Engineer – AI/ML Compiler
Core
Design, develop, and maintain the end-to-end compilation pipeline for Qualcomm AI Hub Workbench, enabling developers to compile PyTorch and ONNX models into deployable artifacts for on-device execution.
Role type
Staff Machine Learning Engineer (AI/ML Compiler)
Builds
Production ML compilation pipelines and tooling for Qualcomm AI Hub
Domain
Hardware engineering, software engineering, systems engineering, and ML infrastructure
Deliverable
production ML models
Required skills
Python, C++, ML compiler concepts (graph IRs, operator fusion, shape inference, lowering passes, backend partitioning), PyTorch model export, ONNXRuntime, CI/CD pipelines
Preferred skills
ML infrastructure experience, compiler stacks (MLIR, ONNX, TVM), SoC-level constraints, hardware-specific runtimes (QAIRT/QNN), automated CI/CD for model compilation
Technologies
PyTorch, ONNX, ONNXRuntime, LiteRT, QAIRT, QNN, MLIR, TVM, torch.export, torch.compile, FX, ATen IR
Responsibilities
Design and maintain the end-to-end compilation pipeline from model ingestion to backend dispatch; Build and maintain ONNX-based and PyTorch compilation paths; Contribute to ONNXRuntime QNN execution provider; Build tooling to analyze, profile, and debug compilation failures; Own compilation and validation of models published on Qualcomm AI Hub; Build and maintain automated compilation pipelines and CI/CD evaluation harnesses
Seniority
Staff, hands-on IC with technical guidance and mentorship
