As a world-renowned VR/AR brand with independent innovation and R&D capabilities, PICO has been at the forefront of the consumer electronic market. We have teams in Europe, Japan and South Korea. Now we are looking for experts in image pipeline to join us to build our AR/VR imaging team.
Responsibilities:
- Design and implement ML compiler for proprietary custom edge NPU architectures that meet latency, memory and power targets.
- Implement operator fusion, memory planning and target lowering passes that support both static and dynamic shape compilation flows.
- Apply knowledge of hardware architecture to optimize latency, memory footprint and bandwidth, most importantly power and thermal constraints.
- Work closely with architecture and runtime engineers to define, develop and debug ML inference on target hardware platforms.
Requirements
Minimum Qualifications
- Master's degree in Computer Science, Electrical Engineering, Computer Engineering, or a related field, or equivalent practical experience.
- 3+ years of industry experience in compiler development, ML systems.
- Solid understanding of compiler fundamentals: IR design, graph transformations, scheduling, memory planning.
- Strong understanding of deep learning model structures of both CNNs and Transformers.
- Understanding of hardware concepts: memory, cache, DMA, tiling, vectorization, systolic array, etc.
- Strong C/C++ or Rust proficiency.
Preferred Qualifications
- 5+ years of relevant industry experience.
- Understanding of quantization concepts (e.g. Observers, static/dynamic quantization).
- Experience with ML compiler stacks such as torch.compile, MLIR, XLA, IREE or TVM.