Research Robotics/Computer Vision Engineer

Responsibilities

• Implement perception on robots to enable safe exploration and navigation in real-world environments
• Reconstruct 3D scenes using monocular images, estimate camera poses, and optimize 3D SLAM
• Develop software tools for robot localization using visual inputs
• Build robust software for life-long mapping via optimally merged pose-graphs
• Implement visual servoing to control robot motion based on detected/tracked objects
• Research novel techniques to mitigate glare during robotic mapping and navigation
• Build infrastructure and pipelines to collect data for training robot manipulation tasks
• Maintain camera and 2D lidar-based navigation stacks, including bug fixes and feature requests

Requirements

• Master’s degree in Computer Vision, Robotics, or a related discipline
• One year of experience in Machine Learning or Data Science
• Experience with 3D scene reconstruction (monocular videos, meshes, pointclouds, NeRF, Gaussian Splats)
• Knowledge of reconstructing rigid and articulated hand-held objects from video
• Experience with generative computer vision and diffusion models for reconstruction
• Ability to optimize attention-based models for autonomous navigation
• Experience using Neural Architectural Search (NAS) for perception backbones
• Proficiency in cloud-based training (AWS, Google Cloud, or Vertex AI) and optimized data loading for distributed deep learning workloads

About the Company

Skild AI is building the world's first general purpose robotic intelligence that is robust and adapts to unseen scenarios without failing through massive scale and data-driven machine learning.