Principal Software Engineer - Compilers
Core
Design and develop scalable model conversion pipelines translating ONNX, TensorFlow, and PyTorch models into an internal Intermediate Representation (IR) for edge AI inference.
Role type
Principal Front-End Compiler Engineer
Builds
Production-ready AI compiler stack and model conversion frameworks for edge AI processors
Domain
Semiconductor / Edge AI / Compiler Engineering
Deliverable
production ML models
Required skills
Compiler design, graph-based optimizations, Python, ONNX, TensorFlow, PyTorch, Intermediate Representations (IRs), graph parsing, graph transformations, pattern matching, graph rewriting, CNNs, Transformers, LLM architectures
Preferred skills
MLIR, TOSA, StableHLO, Torch-MLIR, IREE, TVM, XLA, TensorRT, ONNX Runtime, GPU/NPU/DSP targeting, quantization, deployment workflows
Technologies
Python, ONNX, TensorFlow, PyTorch, MLIR, TOSA, StableHLO, Torch-MLIR, IREE, TVM, XLA, TensorRT
Responsibilities
Design compiler front-end architecture and model conversion frameworks; Build Python-based converters for major ML frameworks; Implement graph construction, transformation, IR lowering, and optimization pipelines; Develop graph optimization passes (operator fusion, canonicalization, simplification, decomposition); Design scalable pattern-matching and graph-rewrite frameworks; Enable support for modern AI architectures (CNNs, Transformers, LLMs); Drive validation, testing infrastructure, regression testing, and CI/CD integration; Debug issues across conversion and optimization stages; Provide technical leadership and mentor engineers
Seniority
Principal, hands-on IC with technical leadership
