Senior Applied Research Engineer - Video
Remote • Europe • FullTime
Posted 21d ago
Remote Work Policy
Fully remote
Employment Type
FullTime
Categories
AI Research Engineer
About the job
Synthesia is seeking an Applied Research Engineer to join their Video team and contribute to building the next generation of production-grade foundation models for human-centric video generation. This role involves working at the intersection of large-scale generative modeling, distributed systems, and production engineering, with a focus on developing and optimizing video base models for realistic, controllable, and expressive synthetic humans. This is an applied research position with direct product impact, requiring the candidate to advance training recipes, scale distributed systems, improve evaluation frameworks, and optimize inference for real-world deployment. The work will directly influence models used by tens of thousands of businesses globally.
Responsibilities
- Own and execute end-to-end research and engineering projects from hypothesis to production impact.
- Develop and scale latent video diffusion models tailored for human-centric video generation.
- Design conditioning mechanisms to improve control (pose, emotion, script, camera) without sacrificing fidelity.
- Advance distributed training strategies (DDP, FSDP, DeepSpeed, sequence parallelism) under real compute constraints.
- Improve training stability at multi-node scale.
- Design rigorous evaluation frameworks combining automated metrics and structured human evaluation.
- Optimize inference for low latency, high resolution, and cost efficiency.
- Run controlled ablations and experiments to drive high-signal modeling decisions.
- Contribute to high engineering standards including reproducibility, experiment tracking, CI/CD, and monitoring.
- Move fast, run multiple hypotheses in parallel, identify signal early, and focus on outcomes.
Requirements
- Strong experience training deep learning models at scale.
- Strong Python and PyTorch skills.
- Hands-on experience with diffusion models (image domain required; video preferred).
- Experience with large scale multi-GPU / multi-node training.
- Good understanding of distributed training (DDP, FSDP, DeepSpeed or similar).
- Ability to design controlled experiments and interpret noisy results.
- Experience with video diffusion models (nice-to-have).
- Experience in avatar or human-centric generation (nice-to-have).
- Familiarity with world / interactive models (nice-to-have).
- Experience with GANs or VAEs (nice-to-have).
- Experience optimizing inference systems for production (nice-to-have).
- Research-driven but outcome-focused.
- Ability to ship products, not just publish research.
- Ability to explore multiple ideas quickly and drop low-signal directions early.
- Clear communication and scientific presentation skills.
- Ability to operate independently but collaborate actively across teams.