About the job
In this role, you will collaborate closely with cross-functional engineering teams to design and deploy state-of-the-art machine learning models directly onto our real-time vehicle platform. This is an exciting opportunity to solve complex, high-impact autonomy challenges and directly shape the future of the Aurora Driver.
Responsibilities
Develop and Optimize Core Perception Solutions: Research and develop state-of-the-art deep learning and machine learning models to improve perception under challenging scenarios, such as long-range multi-sensor detection and degraded sensor conditions.
Tackle End-to-End Autonomy Challenges: Address object detection, tracking of traffic actors, action recognition, and semantic understanding of diverse traffic scenes.
Deploy Production-Ready Software: Guide software development from initial prototype to production deployment on a real-time AV platform, leveraging large-scale data sets for training and analysis.
Collaborate and Iterate: Partner with team members to diagnose, analyze, and resolve failure modes encountered during on-road and simulated testing to produce robust, well-tested systems.
Scale ML Engineering Pipelines: Build and scale robust ML pipelines to facilitate quick experimentation as well as large-scale training and testing.
Qualifications
Minimum
BS, MS, or PhD in Computer Science, Robotics, Engineering, or a related field with a strong foundation in one or more focus areas of ML, including deep learning, computer vision, recursive state estimation, structured prediction, and optimization.
6+ years of research-based or professional experience with C++ and Python.
Comprehensive grasp of linear algebra, discrete/continuous optimization, supervised/unsupervised methods, and generative/discriminative models.
Strong publication record at top-tier robotics/computer vision conferences/journals, or significant industry experience in relevant fields (robotics, computer vision, self-driving technology)
Preferred
Experience utilizing deep learning frameworks such as PyTorch or TensorFlow.
Advanced production-level knowledge of C++ is highly preferred.
Prior experience applying computer vision and machine learning directly to complex robotics problems.
Experience deploying computer vision/ML models at scale across large physical fleets.
Proven ability to work effectively in environments requiring close cross-team collaboration.
Experience working with 3D data, including 3D object detection, tracking, semantic segmentation, or processing point clouds from LiDAR, Radar, and multi-camera system