Synthesia11.03.2026

Senior Research Engineer - Video Foundation Models (Pre - Training)

Полная занятостьУдалёнка

Обязанности

  • 01Developing and scaling latent video diffusion models tailored for human-centric video generation
  • 02Designing conditioning mechanisms to improve control (pose, emotion, script, camera) without sacrificing fidelity
  • 03Advancing distributed training strategies (DDP, FSDP, DeepSpeed, sequence parallelism) under real compute constraints
  • 04Improving training stability at multi-node scale
  • 05Designing rigorous evaluation frameworks combining automated metrics and structured human evaluation
  • 06Optimizing inference for low latency, high resolution, and cost efficiency
  • 07Running controlled ablations and experiments to drive high-signal modeling decisions
  • 08Contributing to high engineering standards: reproducibility, experiment tracking, CI/CD, monitoring

Требования

  • 01Strong experience training deep learning models at scale
  • 02Strong Python and PyTorch skills
  • 03Hands-on experience with diffusion models (image domain required; video preferred)
  • 04Experience with large scale multi-GPU / multi-node training
  • 05Good understanding of distributed training (DDP, FSDP, DeepSpeed or similar)
  • 06Ability to design controlled experiments and interpret noisy results

Условия

  • 01Competitive compensation (salary + stock options + bonus)
  • 02Fully remote from Europe or hybrid work setting with an office in London, Amsterdam