Senior, Computer Vision/Deep Learning Researcher, Advanced Research

Job Description

We have an exciting opportunity in our perception team for individuals with exceptional research and implementation skills in Computer Vision and Deep Learning. The primary responsibility of the Senior, Computer Vision/Deep Learning Researcher is to conduct independent research and develop new core perception technologies within an agreed-upon scope and schedule defined with the management team. Qualified candidates will be driven self-starters, robust thinkers, strong collaborators, and adept at operating in a highly dynamic environment. We look for colleagues that are passionate about our product and embody our values.

Responsibilities

• Conduct independent research and develop state-of-the-art implementations of advanced computer vision capabilities, such as Neural Rendering, 3D Reconstruction, Realtime 6DoF Object Pose Tracking, and 3D Scene Understanding
• Provide leadership and mentorship to more junior software engineers and interns
• Work hand-in-hand with key stakeholders and developers across the company contributing to relevant computer vision capabilities
• Write maintainable, reusable code, leveraging test-driven principles to develop high-quality deep learning and computer vision modules
• Troubleshoot and resolve software defects and other technical issues
• Review development code across the team to ensure high code quality, and reliable results

Qualifications

• 2+ years of working experience in Computer Vision targeted to advanced research which informs and guides future product development
• Expert knowledge in Computer Vision and Deep Learning in the following domains:
• Neural Rendering and Neural Fields: Realtime Novel View Synthesis (NVS); Realtime and Offline Neural 3D Reconstruction (including extraction of 3D geometry, texture, and lighting) from images
• Conditional Generative Modeling: generation and manipulation of 2D images / video, 3D models, and pose trajectories conditioned on textual and visual inputs; strong understanding of image-to-image translation, Diffusion Models, Autoregressive Generative Models, and 3D GANs
• 3D Deep Learning: Realtime 3D Object Detection, 6DoF Object Pose Tracking, and 3D Segmentation; DL-based Feature Extraction and Matching; Point Cloud-based Deep Learning (e.g., PointNet++, Point Transformer, Point NeRF)
• Classical 3D Computer Vision: Strong understanding of classical 3D geometric methods, including Multiple View Geometry, Structure from Motion (SfM), Point Cloud-based inference, camera calibration, feature matching-based registration, and non-linear optimization

• Strong general Deep Learning skills (e.g., 2D Object Detection, Semantic & Instance Segmentation, SSL pre-training, Distillation, DNN architectures, Attention, ViTs, UNets, Multi-task DL, etc.)
• Skilled with both pure / end-to-end Deep Learning and hybrid Classical methods
• Strong knowledge of Python and its ML/CV related ecosystem
• Expert with Pytorch and its extended ecosystem (alternately, JAX, TensorFlow, or other comparable Deep Learning Framework)
• Strong working knowledge of COLMAP and its extensions (or comparable modern, modular SfM pipeline)
• Good working knowledge of C++ (programming and debugging)
• Strong Computer Graphics Rendering background is a plus
• Knowledge of AutoML (including NAS) and Meta / Few-Shot Learning is a plus

Education

• MS in Computer Science, Electrical Engineering or a related field (with a minimum of 3 years of relevant experience)
• Ph.D. is preferred (with a minimum of 1 year of relevant experience)

Additional Information

• All your information will be kept confidential according to Equal Employment Opportunities guidelines