Software Engineer II, Perception

Responsibilities

• Develop end-to-end perception solutions for dynamic-actor tracking, action and attribute recognition, and semantic understanding of traffic environments.
• Research and implement machine learning models for multi-sensor tracking, object classification, and challenging weather or sensor conditions.
• Design algorithms using information fusion, recursive state estimation, Hidden Markov models, structured prediction, and sequential models.
• Build efficient large-scale model training pipelines and analyze failure modes to improve system performance.
• Drive software from prototype through production, including large-scale data analysis and deployment on the real-time Aurora Driver platform.
• Collaborate cross-functionally and contribute to scalable perception-system architecture and testing.

Requirements

• Master’s degree or higher in Computer Science, Robotics, Electrical Engineering, or a related field, or a bachelor’s degree with at least 1 year of relevant industry software-engineering experience.
• Proficiency in C++ and Python for clean, scalable, efficient, production-quality software.
• Basic knowledge of deep learning, computer vision, recursive state estimation, structured prediction, numerical optimization, or related areas.
• Basic knowledge of 3D object-tracking algorithms.
• Experience with PyTorch, TensorFlow, or other deep-learning frameworks.
• Preferred qualifications include VLM, foundational-model, or world-model experience; CUDA or high-performance GPU programming; production-grade real-time CV/ML deployment; ML workflow automation; training-pipeline optimization; open-source contributions; and relevant robotics, computer-vision, or self-driving publications or industry experience.

Benefits

• Eligible for an annual bonus, equity compensation, and benefits.
• Pittsburgh-based roles are expected to work from the office at least 3 days per week, with some roles requiring up to 5 days per week.
• Aurora provides reasonable accommodations for qualified applicants with disabilities and disabled veterans.