Senior Deep Learning Scientist, Multimodal Agentic RL

What you'll do

• Develop, train, fine-tune, and deploy large language models for agentic systems with audio-visual reasoning.
• Advance post-training and alignment methods including instruction tuning, preference optimization, and RLHF/RLVR/MOPD.
• Research and develop agentic reasoning capabilities for planning, tool execution, and long-horizon task completion.
• Lead the collection, development, and benchmarking of high-quality multimodal datasets.
• Evaluate model accuracy, safety, and task completion success across diverse modalities.
• Manage the full model development life cycle, including dataset versioning and experiment tracking.
• Implement deep reinforcement learning algorithms for multimodal agents in complex environments.

What we're looking for

• Master’s degree (or equivalent experience) or PhD in Computer Science, AI, or Applied Math with 5+ years of relevant work experience.
• Excellent programming skills in Python with strong fundamentals in scalable model development and deep learning frameworks like PyTorch.
• Strong knowledge of ML/DL techniques and modern foundation model architectures, including Transformers and mixture-of-experts models.
• Foundational understanding of reinforcement learning algorithms and implementation, including MDPs, policies, and reward design.
• Hands-on experience in post-training multimodal models for omni-modality reasoning, full-duplex voice chat, and human-AI interaction.
• Proven ability to manage model development life cycles, including dataset versioning, experiment tracking, and evaluation pipelines.
• Strong record of publications in top-tier AI and machine learning venues such as NeurIPS, ICML, ICLR, or CVPR (preferred).
• Validated experience training and deploying multimodal foundation models using large-scale distributed infrastructure (preferred).