AI Inference Optimization Engineer
On-site
Core
Deploy and optimize AI models (CNN, LLM, VLM, VLA) for real-time inference on Quadric's General Purpose Neural Processing Unit (GPNPU) for edge devices.
Role type
Senior IC AI Inference Optimization Engineer
Builds
Optimized model deployment flows, profiling/tracing tools, and kernel libraries for GPNPU
Domain
Edge AI, Embedded Systems, Automotive, Robotics
Deliverable
production ML models
Required skills
C/C++, Python, Performance analysis (profilers, roofline), Quantization (PTQ/QAT, int8/int4), Kernel development (CUDA, Triton, OpenCL, SIMD), Model deployment
Preferred skills
Pruning and sparsification, Accelerator microarchitecture, Fixed-point numerics, Compiler code generation
Technologies
GPNPU, CUDA, Triton, OpenCL, Python, C/C++