Join a high‑impact engineering team building a real‑time Edge Unified Road Perception system for automotive and embedded platforms. You’ll work at the intersection of software, hardware and AI, integrating perception models and ensuring performance on edge devices.
What You’ll Do
- Integrate AI perception models into edge and automotive software stacks.
- Create validation and testing pipelines for object detection, segmentation, lane detection, and tracking.
- Measure latency, FPS, accuracy, memory, and GPU/CPU utilization.
- Build automated benchmarking and regression testing frameworks.
- Optimize inference pipelines on NVIDIA Jetson, Qualcomm Snapdragon, and ARM‑based platforms.
- Deploy using TensorRT, ONNX Runtime, CUDA, and OpenCV.
- Troubleshoot integration, runtime, and hardware issues with cross‑functional teams.
What You Need
- Strong C++ and Python programming skills.
- Experience with Linux‑based embedded and real‑time systems.
- Hands‑on deployment of AI models using CUDA, TensorRT, ONNX, OpenCV.
- Proven track record in performance profiling and edge AI optimization.
- Building automated testing, benchmarking, or evaluation tools.
- Knowledge of computer‑vision pipelines and evaluation metrics.
Good to Have
- Experience with automotive perception or ADAS systems.
- Familiarity with YOLO models, multi‑camera pipelines, ROS/ROS2, or GStreamer.
- Deploying AI on automotive or embedded SoC platforms.
The Opportunity
You’ll work alongside AI scientists, embedded engineers, and platform specialists to deliver production‑grade, real‑time road perception solutions. HERE offers a collaborative, inclusive culture that encourages continuous learning and career growth.