Senior AI Algorithm Developer

Stoneridge•Novi, MI

About The Position

The AI Algorithm Developer will work with engineering teams to launch advanced AI features in Stoneridge’s MirrorEye product and other camera-monitor edge systems. The role also involves developing and deploying AI features on hardware-in-the-loop (HiL) systems for demonstrations. The resulting software will be used in production projects for commercial vehicles and heavy machines worldwide. The position focuses on deploying AI features based on computer vision, machine learning, deep learning, image stitching, and object detection and tracking. Additional work includes data collection, system simulation, data analysis, and in-vehicle validation.

Requirements

Master’s degree in Robotics, Electrical Engineering, Computer Science, Mechanical Engineering, or a related field, with strong AI research and development experience in developing ADAS features.
Knowledge and experience in data recording and analysis, requirement analysis, verification, customer interfacing, and cooperation with cross-functional teams.
Coding, modeling, and simulation experience with MATLAB/Simulink, C/C++, and Python.
Experience shipping computer vision and/or sensor fusion systems that fuse multiple data streams, such as camera, IMU, GPS, LiDAR, and wheel odometry, in production environments with high precision and low latency.
At least 3 years of programming experience in C/C++, MATLAB/Simulink, and Python.
At least 3 years of experience working with ML frameworks, libraries, data structures, data modeling, and software architecture.
At least 3 years of experience working with embedded software.
At least 3 years of experience working with OpenVX, OpenCV, OpenCL, and edge ML frameworks.
At least 3 years of experience working on ADAS components using applied machine learning and computer vision.

Responsibilities

Work with algorithm developers and system and software teams to deploy ADAS features.
Design and optimize multimodal ADAS features for commercial vehicles using computer vision, machine learning, and object tracking, integrating camera, IMU, radar, LiDAR, wheel odometry, GPS, and ultrasonic sensor inputs where applicable.
Develop, deploy, and debug real-time, safety-critical embedded software in C/C++ and Python on diverse SoCs and MCUs, including ARM-based platforms with hardware accelerators and GPUs.
Develop RTMaps models for data logging and implement ADAS features in RTMaps for demonstration purposes.
Implement and benchmark computer vision algorithms using OpenCV, OpenVX, OpenCL, and Vulkan on resource-constrained embedded platforms.
Accelerate deep neural network inference for edge deployment using TensorRT, ONNX, and hardware-specific optimization techniques to meet real-time performance constraints.
Prepare technical documentation, validation reports, and calibration procedures; lead technical reviews and contribute to IP generation initiatives.
Contribute to software development compliant with ISO 26262 functional safety and ISO/SAE 21434 cybersecurity standards within an agile framework using Git, JIRA, and CI/CD pipelines.
Work with system and software teams to integrate features and verify that they meet functional requirements on embedded target environments.
Define test methodology using Design of Experiments (DoE), participate in vehicle validation when needed, analyze feature test logs, and create test reports.