Karthik Ragunath Ananda Kumar is an AI Researcher at Tavus, where he focuses on neural rendering and generative modeling to create high-fidelity digital humans capable of real-time, lip-synced interaction. His work spans 3D Gaussian splatting, NeRF architecture variants, and audio-driven avatar synthesis.
Karthik brings deep technical experience from prior roles at Samsung Research America, Ancestry, and Mad Street Den, with research spanning neural texture rendering, generative motion models, and large-scale entity linking.
He holds a master’s degree in Computer Science (Intelligent Systems) from the University of Texas at Dallas, where he researched deep reinforcement learning for autonomous driving.
Outside of Tavus, Karthik is an active contributor to the applied AI community, frequently publishing and presenting on topics like multi-turn LLMs, language-driven avatars, and human-AI interaction design.
