Senior Deep Learning Scientist, Multimodal Agentic RL

Advance streaming and agentic multimodal AI by applying deep learning and reinforcement learning to develop, fine-tune, and deploy language and multimodal models. Work on post-training and alignment methods such as instruction tuning and preference optimization (including RLHF/RLVR), and build agentic reasoning with grounded perception for long-horizon tasks. Lead multimodal dataset collection, benchmarking, and evaluation pipelines for accuracy, safety, and success.