Multimodal Reinforcement Learning Algorithm Researcher
Core
Researching reinforcement learning algorithms for multimodal models, including diffusion and autoregressive models for image/video generation and understanding.
Role type
Research scientist (multimodal reinforcement learning)
Builds
Reinforcement learning training frameworks, reward modeling strategies, and next-generation RL paradigms.
Domain
Artificial Intelligence / Machine Learning / Multimodal Systems
Deliverable
production ML models
Required skills
Reinforcement learning algorithm design, diffusion models, autoregressive models, deep learning system implementation, model training optimization, distributed training, CPU/GPU acceleration
Preferred skills
Text-to-image generation, text-to-video generation, ACM/NOIP participation experience
Technologies
Diffusion models, Autoregressive models, Distributed training frameworks
Responsibilities
Conduct research on RL algorithms for multimodal models; Design and develop RL training frameworks and reward modeling strategies; Explore next-generation RL paradigms for efficient environmental feedback learning.
Seniority
Senior, research-focused IC