AI Application Engineer (Automotive SoC)

About the role

Renesas Electronics is looking for an AI Application Engineer in Kodaira to support the enablement, optimization and deployment of AI models on automotive-grade SoCs. You will work closely with internal compiler and runtime teams and external customers to move AI models from training to optimized inference on embedded NPU/DSP platforms, with a focus on performance, accuracy and system integration.

Key responsibilities

• Enable and deploy AI models (BEV, object detection, segmentation, classification) on Gen4/5 SoC platforms with CNNIP/DSP/NPU HWA, and analyze latency, throughput and multi-core scaling to find bottlenecks in memory bandwidth, scheduling or operator mapping.
• Support post-training quantization (PTQ), quantization-aware training (QAT), operator fusion, graph optimization and execution partitioning, and propose mitigations for accuracy loss.
• Integrate models into Linux/QNX runtimes, debug offloading, memory allocation (IPMMU) and data transfer, and validate workloads on target boards and SIL/HIL simulators.
• Work with AI compiler and runtime toolchains (ONNX-based workflows, hybrid compiler, MWMX) and support ONNX model handling, including graph inspection, segmentation and QDQ models.
• Build or maintain internal tools and scripts for model validation and benchmarking, and act as technical interface for customer evaluations, PoCs and demos.
• Provide technical guidance, documentation and best practices, and contribute to weekly reports, issue tracking and release validation.

Required profile

• Bachelor's or Master's degree in Computer Science, Electrical Engineering, Embedded Systems, or equivalent embedded systems experience.
• Solid understanding of deep learning fundamentals and inference pipelines.
• Hands-on experience with PyTorch, ONNX or ONNX Runtime.
• Strong Python; C/C++ is a plus.
• Familiarity with embedded systems and debugging tools, and ability to analyze performance metrics.
• Good communication in a multi-cultural, cross-functional team.

Preferred qualifications

• 1-3 years in embedded systems or AI development.
• Experience with computer vision models, automotive or robotics use cases, and inference optimization on NPU/DSP/GPU/CPU.
• Knowledge of quantization (INT8, calibration, QDQ) and automotive SoCs or QNX.
• Nice to have: customer-facing application engineering, automotive AI standards, reading and debugging ONNX graphs.