Tavus
AI Researcher (Multimodal Audio/Video Generation)
San Francisco · On-site · Full-time · Posted May 15, 2026
- Experience
- 2–3+ years of hands-on experience applying generative models at scale.
- Education
- PhD or equivalent research experience.
About the job
Leads research and development of conversational audio-visual avatar generation, turning new generative-model research into production systems.
What you'll do
- Lead research on audio-visual avatar generation for conversational settings, including neural avatars and talking heads.
- Develop models that synchronize verbal and non-verbal signals with conversation flow; advance diffusion, long-video generation, and audio-visual modeling.
- Partner with Applied ML and engineering to productionize research, mentor researchers, set research direction, and publish work.
What they ask for
- Expertise in diffusion models and awareness of efficiency techniques.
- Experience in multimodal generation spanning video, audio, and language.
- Proven innovation in long-video generation and/or audio generation.
- Fluent in PyTorch and GPU-optimized workflows; publication record at top-tier venues and experience leading research or mentoring teams.
Nice to have
- Skills in 3D graphics, Gaussian splatting, or large-scale training setups.
- Broad exposure to generative AI models beyond your specialty; familiarity with software development best practices.
Tools
PyTorch
AI focus
Research and develop generative AI for conversational audio-visual avatars, including synchronized verbal and non-verbal signals, diffusion models, long-video generation, and audio-visual modeling.
Model-written summary · the original posting is authoritative