Staff Research Engineer - Multimodal Generative Modelling
This role involves developing advanced multimodal generative models that enable natural, real-time conversational interactions by integrating text, voice, and video. The engineer will lead research and implementation of low-latency, emotionally expressive AI systems, working across pretraining, post-training, and production deployment. Key focus areas include neural codecs, diffusion models, and streaming architectures to enhance realism and interactivity in Synthesia's AI video platform.