Autonomous Driving Software Engineer (Machine Learning Eng.)

ADASTEC delivers SAE Level-4 Automated Driving Software Platform for commercial vehicles to enable OEMs to develop modern, automated, shared, and connected commercial vehicles. ADASTEC flowride.ai SAE Level-4 Automated Driving Software Platform integrated buses are on public roads in 14 countries in Europe & North America. flowride.ai is designed with the fusion of different high-precision sensors available in the global market and complies with safety standards and regulations. ADASTEC has an HQ and operation office in Michigan, USA, an EU Operations office in the Netherlands, Germany, and an R&D Office in Türkiye. For more information, visit www.adastec.com We are hiring a Autonomous Driving Software Engineer who will work as a Machine Learning Engineer, to train, evaluate, and deploy perception models across 2D camera detection, monocular 3D detection, LiDAR 3D detection, and camera–radar fusion; to work with dataset and annotation owners on automated training and labeling, and on watching model drift and domain shift. RESPONSIBILITIES Training and fine-tuning deep learning models for 2D and 3D object detection Developing and evaluating models across camera, LiDAR, and camera–radar fusion Building and maintaining automated training and auto-labeling pipelines Running evaluation, error analysis, and experiment tracking Monitoring model drift and domain shift and feeding findings back into data and training Exporting and validating models for onboard deployment QUALIFICATIONS Bachelor’s or master’s degree in computer science, AI, Electrical & Electronics Engineering, Robotics, or a related field Excellent command of English, both written and spoken Minimum 3 years of professional experience as a Machine Learning Engineer Strong understanding of machine learning and deep learning fundamentals Hands-on PyTorch experience with model training, validation, and evaluation Experience with object detection and transfer learning Comfort with at least two of: 2D detection, monocular 3D, LiDAR 3D, multi-modal fusion Good understanding of evaluation metrics and performance analysis Python, Linux, Git Excellent command of English, both written and spoken. 2D and 3D object detection on camera, LiDAR, or radar Camera–radar or other multi-sensor fusion Experiment tracking (W&B, MLflow, or similar) Model export and optimization (ONNX, TensorRT, or similar) Dockerized GPU training and distributed training Automated training or labeling pipelines Domain shift / model drift monitoring ROS / ROS2 familiarity C++ is a plus Having no travel restriction.