Lead - ADAS Computer Vision / Perception
Role Summary :
We are looking for a senior Computer Vision / Perception expert for ADAS perception team. This role will be responsible for developing camera-based perception algorithms, building sensor fusion capabilities, and creating scalable pipelines that enable advanced driver assistance features. The candidate will act as the foundational technical leader, driving both hands-on development and team building.
Key Responsibilities :
Computer Vision & Perception Development :
- Design and implement perception algorithms for object detection, tracking, lane detection, drivable area segmentation, and free space estimation.
- Develop multi-sensor fusion (camera + radar + LiDAR optional) for robust perception.
- Optimize algorithms for real-time performance on automotive-grade hardware (NVIDIA, Qualcomm, Renesas, TI etc.).
- Ensure perception pipeline robustness under diverse lighting, weather, and road conditions.
System & Integration :
Define perception requirements and KPIs aligned with ADAS features and NCAP regulations.Work closely with planning & control teams to ensure perception outputs meet latency and accuracy needs.Integrate with embedded software frameworks (ROS2, AUTOSAR, middleware).Collaborate with vehicle integration engineers for sensor calibration and validation.Leadership & Roadmap :
Build and mentor a team of computer vision and ML engineers.Define a scalable perception roadmap (from L1 / L2 to L2+ features).Drive data collection, labeling, and dataset curation strategies.Collaborate with external partners, Tier-1 suppliers, and research labs.Required Qualifications :
710 years of experience in computer vision for ADAS, robotics, or autonomous driving.Strong expertise in deep learning for perception (detection, segmentation, tracking).Hands-on experience with CNNs, transformers, multi-task learning (PyTorch / TensorFlow).Background in sensor calibration, camera models, and 3D geometry.Experience deploying ML models on embedded / edge hardware (TensorRT, ONNX, OpenVX).Strong programming skills in C++ and Python; exposure to CUDA / OpenCL.Familiarity with automotive development processes (ISO 26262, ASPICE, SOTIF).Preferred / Bonus Skills :
Experience with multi-sensor fusion (camera + radar / LiDAR).Familiarity with SLAM, HD maps, or localization.Prior experience in ADAS production programs (e.g., NCAP AEB, LKA).Simulation & validation experience (CARLA, IPG CarMaker, dSPACE).Exposure to active safety testing protocols (Euro NCAP, US NCAP, AIS).(ref : hirist.tech)