Senior/Staff Machine Learning Engineer, Perception
South San Francisco, CAFull-TimeStaffAI / Data Science
What you'll do
- Develop computer vision and machine learning models for real-time perception systems, enabling tractors to identify crops, obstacles, and terrain in varying unpredictable conditions.
- Build sensor fusion algorithms to combine camera, LiDAR, and radar data, creating robust 3D scene understanding that handles challenges like crop occlusions or GNSS drift.
- Optimize models for low-latency inference on resource-constrained hardware, balancing accuracy and performance.
- Design and test data pipelines to curate and label large sensor datasets, ensuring high-quality inputs for training and validation, with tools to visualize and debug failures.
- Analyze performance metrics and iterate on algorithms to improve accuracy and efficiency of various perception subsystems.
What you’ll bring
- A MS, or PhD in Computer Science, AI, or a related field, or 5+ years of industry experience building vision-based perception systems.
- Deep expertise in developing and deploying machine learning models, particularly for perception tasks such as object detection, segmentation, mono/stereo depth estimation, sensor fusion, and scene understanding.
- Strong understanding of integrating data from multiple sensors like cameras, LiDAR, and radar.
- Experience handling large datasets efficiently and organizing them for labeling, training and evaluation.
- Fluency in Python and experience with ML/CV frameworks like TensorFlow, PyTorch, or OpenCV, with the ability to write efficient, production-ready code for real-time applications.
- Proven ability to design experiments, analyze performance metrics (e.g., mAP, IoU, latency), and optimize algorithms to meet stringent performance requirements in dynamic settings.
- An eagerness to get your hands dirty and agility in a fast-moving, collaborative, small team environment with lots of ownership.
What makes you a strong fit
- Experience architecting multi-sensor ML systems from scratch.
- Experience with Foundational models for robotics or Vision-Language-Action (VLA) models
- Experience with compute-constrained pipelines including optimizing models to balance the accuracy vs. performance tradeoff, leveraging TensorRT, model quantization, etc.
- Experience implementing custom operations in CUDA.
- Publications at top-tier perception/robotics conferences (e.g. CVPR, ICRA, etc.).
- Passion for sustainable agriculture and securing our food supply chain.
