Senior Deep Learning Scientist, Multimodal Agentic RL

Tasks

• Advance post training and alignment methods
• Collect develop and benchmark multimodal datasets
• Develop train fine tune deploy large language models for agentic systems
• Evaluate model accuracy safety and task success
• Implement reinforcement learning algorithms and reward design
• Manage dataset versioning experiment tracking and evaluation pipelines
• Plan tool execution and long horizon task completion
• Research agentic reasoning and grounded perception

Perks/Benefits

• Comprehensive benefits package
• Equity

Skills/Tech-stack

Dataset versioning | Deep learning | Distributed Training | Experiment tracking | Instruction Tuning | MDP | Mixture of Experts | Policy Optimization | Preference optimization | PyTorch | Python | RLHF | Reinforcement Learning | Reward Design | Transformers