About the Role
Role details below.
Responsibilities
- Develop and train deep learning models for camera-based perception, enabling the autonomy stack to detect objects, understand scenes, and estimate geometric information from visual inputs.
- Implement production-quality machine learning code to support model training, evaluation, and inference for camera perception systems.
- Analyze model performance across diverse driving scenarios, identify failure modes, and improve robustness and generalization.
- Contribute to the development and optimization of large-scale training pipelines, including dataset preparation, distributed training, and experiment management.
- Work closely with data teams to curate and improve training datasets derived from fleet logs, simulation, and annotation pipelines.
- Collaborate with cross-functional teams across perception, simulation, and validation to evaluate model performance and support integration into the autonomy stack.
- Improve experimentation workflows and tooling to accelerate model iteration, reproducibility, and evaluation.
- Contribute to discussions on model architecture, training strategies, and perception system design.
Requirements
- Bachelor’s degree in Computer Science, Robotics, Electrical Engineering, Machine Learning, or a related technical field with 4+ years of industry experience, or a Master’s degree with 2+ years of experience.
- Experience developing machine learning or deep learning models for computer vision or perception systems.
- Strong programming skills in Python and PyTorch, with experience writing production-quality ML code.
- Experience training and evaluating machine learning models using large datasets and scalable compute environments.
- Understanding of modern deep learning architectures used in perception systems, such as CNNs, transformers, or multi-task learning models.
- Experience debugging model behavior, analyzing performance metrics, and iterating on training pipelines.
- Ability to collaborate with cross-functional teams to integrate ML models into larger software systems.
Nice to Have
- Experience working in autonomous driving, robotics, or simulation-based training environments.
- Experience with multi-task learning or perception architectures that combine multiple visual tasks.
- Experience working with large-scale data pipelines, distributed training systems (e.g., Ray), or experiment management frameworks.
- Familiarity with camera calibration, geometric reasoning, or 3D perception from images.
- Experience deploying ML models into production or real-world robotics systems.
