
Responsibilities
-
Design end-to-end differentiable systems that jointly optimize sensor control, illumination patterns, and perception tasks (detection, segmentation)
-
Develop domain adaptation techniques enabling synthetic-to-real transfer without retraining.
-
Implement energy-aware AI models balancing perception quality with power constraints for embedded automotive deployment.
-
Validate systems through real-world vehicle testing across diverse environmental conditions.
-
Publish research at top-tier conferences (CVPR, ICCV, NeurIPS) while delivering production-ready solutions.
Qualifications
-
Bachelor's or Master's degree in Computer Science, Computer Engineering, Data Science, Electrical Engineering, or a related field.
-
3+ years developing and deploying deep learning and reinforcement learning models
-
Expert-level deep learning (PyTorch/TensorFlow) with custom architecture design.
-
Strong background in object detection, semantic segmentation, neural rendering.
-
Experience with domain adaptation, self-supervised learning, and model optimization for
-
edge devices.
-
Familiarity with ROS, simulation environments (CARLA, Applied Intuition, NVIDIA), and 3D
-
rendering engines.
-
Experience with automotive perception datasets (nuScenes, Waymo Open, KITTI) and
-
benchmarks.
-
Hands-on experience with automotive sensors (cameras, LiDAR, radar, thermal imaging).
-
Knowledge of closed-loop control integration with neural networks.
-
Proficiency in Python, C++, and CUDA programming.
© 2026 Qureos. All rights reserved.