Research Scientist - Interactive Avatars

About the role

As a Research Scientist, you will join a team of researchers and engineers working on the frontier of generative AI with a focus on avatar-centric, interactive video diffusion models. You will build models that hold natural, real-time conversations, understanding user input and responding with appropriate turn-taking, gaze, and facial reactions. This role involves turning breakthrough research into product capabilities that ship.

Responsibilities

• Contribute to the research direction for dyadic interaction modeling and own well-scoped research problems end to end.
• Advance the state of the art in the perceptual layer of interactive agents, including understanding user audio and video and generating contextually appropriate reactions.
• Post-train multimodal models to generate rich, natural dyadic interactions from user audio and video inputs.
• Adapt diffusion models to new conditioning signals, such as conversational state, turn-taking, and listener cues.
• Build robust evaluation frameworks and test suites for continuous tracking of interaction quality.
• Partner with our data team to define data needs and shape high-quality datasets.
• Run rigorous experiments and share findings that inform the team's technical decisions.

Required skills

• Machine Learning
• Diffusion models
• PyTorch
• Video generation

Nice to have

• Real-time generation
• Streaming generation
• Autoregressive video diffusion
• Low-latency inference
• Audio-driven motion modeling
• Conversational modeling
• Mentoring

Qualifications

• Publications at top-tier venues such as CVPR, ICCV, ECCV, NeurIPS, ICML, ICLR, or SIGGRAPH
• Equivalent demonstrated impact

About the Company

Synthesia is the world’s leading AI video platform for business, used by over 90% of the Fortune 100. Founded in 2017, the company develops products to enhance visual communication and enterprise skill development.