Staff Research Scientist- Interactive Avatars

Overview

As Staff Research Scientist in Synthesia’s Interactive Avatars team, you define the research direction for dyadic interaction modeling and push the state of the art in multimodal diffusion-based avatar video. You translate breakthroughs into product capabilities that enable natural, real-time conversations with avatars. You work within a fast-moving R&D group to shape scalable, user-facing AI agents while upholding Safety, Ethics and Security norms. This role offers impact across research and product, in a collaborative, aspiration-driven environment.

Pay / Benefits

• Competitive compensation
• Hybrid work setting with offices in London, Zurich, Munich or remote in Europe
• 25 days of annual leave + public holidays
• Strong company culture with regular planning and socials
• Location-based benefits

Responsibilities

• Set the research roadmap for dyadic interaction modeling and balance long-term bets with near-term product impact
• Advance perceptual capabilities for interactive agents, including user audio/video understanding and responsive reactions
• Post-train multimodal models to generate natural dyadic interactions from audio/video inputs
• Adapt diffusion models to conditioning signals like conversational state and listener cues
• Build robust evaluation frameworks and test suites to track interaction quality
• Collaborate with data team to define data needs and curate high-quality datasets
• Lead and mentor a small research team and drive cross-team technical decisions in research, data, and engineering

Key requirements

• Deep ML expertise with diffusion models, preferably for video or avatar generation
• Strong publication record at top-tier venues or demonstrable impact in related work
• Experience leading a small research team and mentoring junior members
• Track record of taking research from idea to production
• Proficiency in PyTorch and modern ML tooling for large-scale training
• Clear communication of hypotheses, experiments, and results; ability to influence direction across teams

• Clear communication
• Leadership and mentoring
• Cross-team collaboration
• Diffusion models for video/avatar generation
• Multimodal modeling (audio/video)
• Post-training model adaptation

Full job description from WhatJobs. Apply for this role →