Research Scientist - Spatial AI & 3D Reconstruction

Mecka; New York (NY); On-site; USD 200,000 - 250,000; Full time

Job summary

Mecka AI is hiring a Research Scientist to architect and train foundation models for 3D spatial and temporal perception. You will build in-house 3D reconstruction tools and robust optical flow models for egocentric vision, enabling advanced perception for embodied AI.

You will work with massive proprietary data, train end-to-end perception loops, and push the frontier in dense reconstruction and neural rendering within a fast-changing robotics domain.

Qualifications

• Deep expertise in Deep Learning, 3D Computer Vision, and Temporal/Video Modeling.
• Proven experience training large-scale vision models from scratch, not just running inference or fine-tuning.
• Strong theoretical and practical understanding of modern feed-forward 3D networks and dense motion estimation.
• Mastery of PyTorch and deep learning scaling frameworks.
• Experience handling and curating massive, multi-terabyte image and video datasets for training.

Responsibilities

• Architect proprietary spatial models from scratch, including 3D geometry extraction architectures.
• Scale multi-view ML architectures across multi-GPU clusters for large spatial datasets.
• Develop novel loss functions and attention mechanisms for data distributions.

Skills

Deep Learning

3D Computer Vision

Temporal/Video Modeling

PyTorch

Large-scale training

Multi-terabyte datasets

CUDA kernels

Tools

CUDA

Job description

Mecka AI is hiring a Research Scientist to architect and train foundation models for 3D spatial and temporal perception. You will build in-house 3D reconstruction tools and robust optical flow models for egocentric vision, enabling advanced perception for embodied AI.

You will work with massive proprietary data, train end-to-end perception loops, and push the frontier in dense reconstruction and neural rendering within a fast-changing robotics domain.