PE- CPU Software & Hardware Co-Design Engineer (ML Systems)
Core
Design CPU software-hardware co-design for next-generation QMX architectures, focusing on ML workload characterization, simulation, kernel optimization, and driving architectural insights for efficient ML execution on CPU platforms.
Role type
Senior IC CPU software & hardware co-design engineer (ML Systems)
Builds
Optimized ML kernels and libraries for QMX architecture, simulation tooling for workload analysis, and data-driven insights for future CPU designs.
Domain
Semiconductor / CPU Architecture / Machine Learning
Deliverable
production ML models
Required skills
C/C++, Computer Architecture, Systems Programming, Machine Learning fundamentals, Performance profiling, Benchmarking, Optimization, SIMD/vectorization, Cache-aware execution, Parallel execution strategies
Preferred skills
QEMU or equivalent simulators, ML kernel development (GEMM, convolution, attention), CPU architecture (pipelines, caching, SIMD/vector extensions), Low-level optimization (intrinsics, assembly, memory tuning), ML frameworks and inference stacks
Technologies
QEMU, PyTorch, ONNX, XNNPACK, MLAS, Geekbench AI
Responsibilities
Identify and prioritize critical ML use cases and models for CPU-centric execution; Generate detailed execution traces for ML workloads using simulators; Identify system bottlenecks across CPU pipelines, memory hierarchy, and instruction utilization; Collaborate with CPU architecture teams to provide data-driven insights and propose architectural enhancements; Design and implement highly optimized ML kernels and libraries for QMX architecture; Optimize CPU-centric ML benchmarks and establish performance baselines.
Seniority
Senior, hands-on IC
