Application Engineer
Core
Enable, optimize, and deploy AI models (BEV, object detection, segmentation) on automotive-grade SoCs with CNNIP/DSP/NPU for embedded inference.
Role type
AI Application Engineer (Embedded Systems)
Builds
Optimized AI inference pipelines on Gen4/5 SoC platforms for automotive applications
Domain
Automotive embedded systems, AI inference, SoC architecture
Deliverable
production ML models
Required skills
Deep learning fundamentals, AI frameworks (PyTorch, ONNX, ONNX Runtime), Python programming, embedded systems debugging, performance analysis (latency, throughput), quantization techniques (PTQ, QAT, INT8), memory hierarchy understanding, multi-core scheduling
Preferred skills
C/C++ programming, automotive SoC experience, QNX environment familiarity, computer vision model training/evaluation, NPU/DSP/GPU inference optimization, DMA knowledge
Technologies
ONNX, ONNX Runtime, Linux, QNX, CNNIP, DSP, NPU, HWA, MWMX, hybrid compiler
Responsibilities
Deploy AI models on Gen4/5 SoC platforms, perform model performance analysis and bottleneck identification, support model optimization workflows (PTQ, QAT, operator fusion), integrate AI models into embedded runtime environments, debug offloading and memory allocation issues, validate workloads on target boards and simulators, develop internal tools for model validation and benchmarking, act as technical interface for customer evaluations and PoCs
Seniority
Mid-level, hands-on IC
