Principal AI Compiler Engineer
Core
Architect and scale a production-class AI compiler toolchain to automate, optimize, and accelerate neural network deployment on NXP's System-on-Chip (SoC) platforms.
Role type
Principal AI Compiler Engineer
Builds
AI compiler toolchain, graph optimization frameworks, and accelerator backends for embedded intelligence.
Domain
Semiconductor hardware design, embedded AI, compiler technology.
Deliverable
production ML models
Required skills
Graph optimization, performance optimization for NPUs/custom accelerators, MLIR/LLVM/TVM-like systems, C/C++, Python, AI inference workloads (CNNs, transformers), AI agent integration
Preferred skills
Quantization, operator fusion, memory planning, layout transformations, AI-enabled profiling and diagnostics, validation and benchmarking pipelines
Technologies
PyTorch, TensorFlow, ONNX, MLIR, LLVM, TVM
Responsibilities
Design and evolve AI compiler toolchain leveraging AI agents; pioneer graph transformations and codegen strategies; build integrations with AI frameworks; optimize quantization and memory planning; co-design compiler features with hardware architects; diagnose performance bottlenecks; uplevel validation and regression pipelines; streamline developer experience and documentation
Seniority
Principal, hands-on IC with strategy & mentorship
