Video Generation Foundation Model Researcher

HOK

Diffusion models
Autoregressive generation
Pytorch
Design and iterate on next-generation video generation foundation architecture

Job Summary

  • Design and iterate on next-generation video generation foundation architecture.
  • Explore core technologies for long video generation, including long-context attention and memory mechanisms.
  • Lead video pre-training at the 10-billion/trillion token scale and define data mixture strategies.

Matching Summary

Design and iterate on next-generation video generation foundation architecture.

Skills & Requirements

Must-have

  • diffusion models
  • autoregressive generation
  • PyTorch
  • large-scale distributed training
  • video tokenizers

Nice-to-have

  • industry product R&D
  • end-to-end video foundation model projects

Key Requirements

  • Ph.D. in AI-related fields
  • First-author papers at top-tier conferences
  • Experience training video/image generation models from scratch
  • 3+ years of relevant research experience
  • Publications related to video generation or diffusion models

Work Rights

Not specified

Tailored Resume

Cover Letter