Edge ML Software Engineer (Compiler-PICO) - San Jose

Build and optimize an ML compiler for PICO’s proprietary custom edge NPU architectures, meeting latency, memory, and power targets. You’ll implement compiler passes such as operator fusion, memory planning, and target lowering for static and dynamic shape compilation, then apply hardware knowledge (cache, DMA, tiling, vectorization, systolic arrays) to meet power and thermal constraints. Collaborate with architecture and runtime engineers to define, develop, and debug ML inference on target platforms.