Generative AI - 3D Foundation Model

Workman Labs; Sunnyvale, Northern (CA, KY); On-site; USD 150,000 - 230,000; Full time

Job summary

MeshyAI in Sunnyvale, CA, is seeking a Senior Researcher to advance 3D-native generative foundation models and translate research into products used by millions. You will design and train large-scale 3D models, bridging research and product with user feedback, and publishing breakthroughs via demos or open-source.

You will collaborate on scalable training pipelines and maintain reproducible experiments in a fast-paced, interdisciplinary team.

Qualifications

• Strong Python and PyTorch experience for research and production.
• Deep understanding of transformer architectures and modern generative models.
• Proven ability to work on 3D content generation and perception.
• Effective communication and collaboration in cross-functional teams.
• Open-source contributions and top-tier conference publications are a plus.

Responsibilities

• Design, train, and refine large-scale 3D generative models across training stages.
• Bridge research and product by deploying models in real applications with human feedback.
• Develop novel architectures for faster, higher-quality 3D generation.
• Collaborate on scalable training and data pipelines across GPU clusters.
• Share breakthroughs via demos, open-source, or technical reports.

Skills

Python

PyTorch

Transformer models

3D generation intuition

Communication

Open-source contributions

Tools

GPU clusters

Experiment tooling

Job description

MeshyAI is hiring a researcher to develop core 3D-native generative foundation models, advancing the state of the art in generative AI for 3D content and translating research into products used by millions.

Location

Sunnyvale, CA

On-site

Seniority

Senior

Compared with 9 other Research roles: Common across similar roles Unique to this role

What you'll do

Responsibilities

• Design, train, and refine large-scale 3D generative models across pre-training, post-training, and emerging paradigms (diffusion, flow matching, multi-modal learning) Common
• Bridge research and product by deploying models in real applications with human feedback
• Develop novel architectures for faster, higher-quality, controllable 3D generation
• Collaborate on scalable training and data pipelines across GPU clusters Common
• Maintain engineering discipline with clean code and reproducible experiments Unique
• Share breakthroughs via demos, open-source, or technical reports

What you'll need

Experience

• Strong Python and deep learning framework skills (preferably PyTorch)
• Deep understanding of transformer architecture and modern generative models Common
• Intuitive grasp of how models perceive and generate 3D worlds Unique
• Self-directed, team-oriented mindset with clear communication Unique
• Plus: high-performance training experience, open-source contributions, top-tier conference publications Common