About the Team
Our team focuses on advancing computer vision and deep learning technologies to enhance perception systems for autonomous systems. We work on state-of-the-art neural network models for LiDAR-to-camera calibration, 3D scene reconstruction, and depth estimation. By combining synthetic data generation, large-scale dataset processing, and rigorous experimentation, we push the boundaries of AI-driven perception. Our work directly contributes to improving the accuracy and reliability of autonomous navigation systems.
About the Role
As a Software Engineer, you will develop and optimize state-of-the-art models for solving real-time perception tasks based on data from various sensors, including lidars, cameras, and radars. You will continuously study and apply cutting-edge research papers focused on perception in autonomous driving. In addition to model development, you will build data processing pipelines, evaluate model performance, optimize inference, and apply innovative solutions—including large multimodal models—to enhance our autonomous driving system. Strong skills in computer vision, deep learning, Python, and C++ are essential.
What You'll Do
- Research and implement state-of-the-art machine learning technologies to build multimodal sensor fusion architectures for Perception tasks (e.g., object detection, scene segmentation)
- Develop deep learning models, including vision language models (VLMs)
- Design and implement data mining, labeling, training and evaluation pipelines
- Develop and extend the Perception library (including the codebase for processing sensor data)
- Identify and research opportunities to optimize the performance of Perception models by analyzing inference traces
- Apply optimizations to the ML model codebase by evaluating the correctness of changes, including retraining the models to evaluate accuracy
- Integrate new models and solutions into the autopilot pipeline and simulation systems
What You'll Need
- MS in Computer Science, Electrical Engineering, Mathematics, or a related discipline; or equivalent practical experience. Foreign equivalent is accepted.
- 5+ years experience in Computer Vision and Machine Learning with hands-on experience building, integrating, debugging and evaluating ML-based systems; corresponding experience with deep learning frameworks such as PyTorch
- 2+ years experience in model optimization or efficient deep learning techniques
- 2+ years experience with sensor data processing (Lidar, camera, Radar)
- 1+ year experience with Vision Language Models (VLM)
- 2+ years experience with autonomous vehicles or ADAS systems (L2/L3/L4)
- Proficiency in Python and C++ for developing and optimizing production-quality ML systems; hands-on experience with managing resources to get the maximum performance out of the system (CPU, GPU, memory, threads)
Nice to Have
Candidates are required to be authorized to work in the U.S. The employer is not offering relocation sponsorship, and remote work options are not available.
