Research Scientist - Spatial AI & 3D Reconstruction

Mecka

New York (NY)

Vor Ort

USD 150.000 - 230.000

Vollzeit

14 Tage+

Erhalte eine Antwort von diesem Arbeitgeber

Dein Lebenslauf und dein Anschreiben, genau auf das zugeschnitten, was gesucht wird.

Schaffe es an den ATS-Filtern vorbei

Zusammenfassung

Mecka is hiring a Research Scientist to architect and train proprietary foundation models from scratch, focusing on 3D geometry extraction and robust optical flow for egocentric vision. You will leverage massive proprietary data to train networks that surpass public baselines and own the spatial-temporal perception loop for our data engine.

You will lead fast-paced R&D efforts, prototype new architectures, and integrate outputs into dense reconstruction pipelines while collaborating across

Qualifikationen

  • Deep expertise in Deep Learning, 3D Computer Vision, and Temporal/Video Modeling.
  • Proven experience training large-scale vision models from scratch.
  • 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.
  • Comfortable operating in a fast-paced environment where priorities can shift rapidly to capitalize on new research or hardware capabilities.
  • Warning: Research Scientist positions require hyper-specific expertise; limit applications to one research role.

Aufgaben

  • Architect proprietary spatial models from zero to one for 3D geometry extraction.
  • Scale multi-view ML architectures across multi-GPU clusters for massive spatial datasets.
  • Develop novel loss functions and attention mechanisms for data-driven paradigms.
  • Build egocentric optical flow models for extreme motion, occlusions, and rapid rotations.
  • Segment dynamic actors and track objects using flow models for downstream world models.
  • Prototype and deploy models for tracking, segmentation, and multi-modal sensor fusion.
  • Integrate model outputs into dense reconstruction pipelines (3D Gaussian Splatting, NeRFs) for photorealistic environments.

Kenntnisse

Deep Learning
3D Vision
Video Modeling
Large-Scale Training
PyTorch
Multi-GPU
Dense Motion Estimation
CUDA

Tools

CUDA
Distributed Training

Jobbeschreibung

Mecka is hiring a Research Scientist to architect and train proprietary foundation models from scratch, focusing on 3D geometry extraction and robust optical flow for egocentric vision. You will leverage massive proprietary data to train networks that surpass public baselines and own the spatial-temporal perception loop for our data engine.

You will lead fast-paced R&D efforts, prototype new architectures, and integrate outputs into dense reconstruction pipelines while collaborating across

Hol dir deinen kostenlosen, vertraulichen Lebenslauf-Check.

oder ziehe deine Datei hierhin.

Similar jobs

Ähnliche Jobs, die dir auch gefallen könnten

Research Scientist — 3D Vision & Egocentric Motion
Research Scientist — 3D Vision & Egocentric Motion

Mecka • San Francisco (CA)

Vor Ort
USD 98.000 - 140.000
Research Scientist: 3D Vision & Egocentric Motion
Research Scientist: 3D Vision & Egocentric Motion

Kindredventures • New York (NY)

Vor Ort
USD 150.000 - 210.000
Research Scientist — 3D Reconstruction & Neural Rendering
Research Scientist — 3D Reconstruction & Neural Rendering

Mecka • New York (NY)

Vor Ort
USD 141.000 - 162.000
Neural 3D Object Reconstruction Scientist
Neural 3D Object Reconstruction Scientist

Mecka • New York (NY)

Vor Ort
USD 141.000 - 162.000
Research Scientist (Spatial AI & Neural Reconstruction)
Research Scientist (Spatial AI & Neural Reconstruction)

Mecka • New York (NY)

Vor Ort
USD 150.000 - 230.000
Research Scientist – 3D Environment Reconstruction
Research Scientist – 3D Environment Reconstruction

Mecka • New York (NY)

Vor Ort
USD 141.000 - 162.000
Research Scientist, Multimodal Foundation Models
Research Scientist, Multimodal Foundation Models

Kindredventures • San Francisco (CA)

Vor Ort
USD 180.000 - 300.000
Research Scientist: 3D Hand Tracking & HOI Architect
Research Scientist: 3D Hand Tracking & HOI Architect

Kindredventures • New York (NY)

Vor Ort
USD 120.000 - 180.000
Lead 3D Reconstruction & Spatial AI Scientist
Lead 3D Reconstruction & Spatial AI Scientist

Tera Ai • San Francisco (CA)

Vor Ort
USD 150.000 - 210.000
Research Scientist- SLAM & VIO
Research Scientist- SLAM & VIO

Mecka • San Francisco (CA)

Vor Ort
USD 98.000 - 140.000