Responsibilities
- Investigate Large Multimodal Models as they relate to Conversational Avatars, including Neural Avatars and Talking-Heads.
- Design systems that capture both spoken and non-spoken elements of dialogue, enabling real-time, low-latency control of avatar behavior.
- Explore fine-tuning, model adaptation, and conditioning strategies to enhance expressiveness, precision, and task alignment in AudioVisual Multimodal Models.
- Collaborate with the Applied ML team to transition experimental models into scalable production systems.
- Monitor emerging research and contribute to shaping the future direction of conversational AI technologies.
Work Arrangement
Hybrid — San Francisco, London
