3 months ago
San Mateo, CA, USAMid Level
H1B Sponsor
Responsibilities
- Develop and optimize deep learning models for depth estimation, object detection, segmentation, tracking, and 3D scene understanding using multimodal sensor data.
- Build scalable data processing, training, evaluation, and deployment pipelines for real-world and real-time systems.
- Design labeling strategies and tooling for automated annotation, quality assurance workflows, dataset management, augmentation, and versioning.
- Implement monitoring and reliability frameworks for uncertainty estimation, failure detection, and automated performance reporting.
- Run proof-of-concept experiments and translate research findings into practical perception prototypes.
- Collaborate with robotics, systems, and simulation teams to integrate perception models into production pipelines and improve end-to-end performance.
Requirements
- Strong experience with deep learning frameworks such as PyTorch, TensorFlow, or JAX.
- Background in computer vision tasks including detection, depth estimation, segmentation, tracking, or 3D scene understanding.
- Proficiency in Python; familiarity with C++ is a plus.
- Experience building training pipelines, evaluation frameworks, and machine learning deployment workflows.
- Knowledge of 3D geometry, sensor processing, or multi-sensor fusion involving RGB-D, LiDAR, or stereo.
- Experience with data annotation tools, dataset management, and augmentation techniques.
- Familiarity with robotics, simulation environments such as Isaac Sim, Gazebo, or Blender, or real-time systems.
- Understanding of uncertainty modeling, reliability engineering, or ML monitoring and MLOps practices.
Categories
ML EngineeringRobotics
About Skild AI
Building general purpose robotic intelligence.