感知模型剪枝量化工程师_XC
Core
Design and implement pruning and quantization strategies for perception models to enable efficient deployment on compute-constrained automotive platforms.
Role type
Senior IC machine-learning engineer (model compression)
Builds
Pruned and quantized perception models for autonomous driving/ADAS systems
Domain
Autonomous driving / ADAS / Embedded systems
Deliverable
production ML models
Required skills
Pruning (structured/unstructured), Quantization (PTQ/QAT), INT8/FP16, Model calibration, Sensitivity analysis, Mixed precision strategies, CNN/Transformer architectures, Object detection/Segmentation/BEV models, PyTorch, TensorRT, ONNX Runtime, Python, C++, Model export/conversion, Performance profiling
Preferred skills
Sparse training, Knowledge distillation, Low-bit quantization, NPU/BPU/GPU optimization, Embedded platform deployment, Automotive mass production delivery
