Perception & Mapping Intern
Aurora
San Francisco, CAFresher13m agovia LinkedIn
About the Role
Aurora's mission is to deliver the benefits of self-driving technology safely, quickly, and broadly. The Aurora Driver will create a new era in mobility and logistics, one that will bring a safer, more efficient, and more accessible future to everyone. At Aurora, you will tackle massively complex problems alongside other passionate, intelligent individuals, growing as an expert while expanding your knowledge.
We're searching for Summer 2027 Perception & Mapping Interns. You'll work on the models that let the Aurora Driver see and understand the world — object detection and tracking, sensor fusion, semantic segmentation, VLMs, environmental/sensor conditions, and the ML-based map models and data that ground the system.
Key Responsibilities
- Develop onboard and/or offboard perception or mapping models to improve autonomous vehicle performance
- Process sensor data (camera, radar, lidar) to detect and track objects, actors, signs, traffic lights, and hazards
- Analyze, root-cause, and help resolve system deficiencies such as false positives/negatives and misclassifications
- Work with large real-world datasets across training, evaluation, and visualization pipelines
- Partner with full-time engineers to contribute to the next generation of the perception or mapping system
Requirements & Tech Stack
- Currently pursuing a Master's or PhD in Computer Science, Robotics, Electrical Engineering, or a related field
- Strong ML and computer vision fundamentals
- Hands-on deep learning experience in PyTorch (or comparable), including model training and debugging
- Proficiency in Python and C++
- Comfort working with large datasets across training, evaluation, and debugging pipelines
- Has built and trained a perception or mapping model end-to-end, not only fine-tuned an existing backbone
Desirable Qualifications
- Experience with VLMs or other pretrained foundation models
- Prior publication record at a top-tier venue
- Knowledge of object tracking, Kalman filtering, or Hidden Markov Models