AI Research Engineer (Agentic Reinforcement Learning) - 100% Remote Worldwide
Core
Drive innovation in multi-modal reinforcement learning to advance next-generation intelligent systems for real-world challenges.
Role type
Senior IC AI Research Engineer (Agentic Reinforcement Learning)
Builds
Next-generation intelligent systems using multi-modal RL, diffusion-based generative models, and autoregressive models.
Domain
Artificial Intelligence / Reinforcement Learning / Multimodal Systems
Deliverable
production ML models
Required skills
Multi-modal reinforcement learning, Advanced model architectures, RL infrastructure design, Reward modeling strategies, Multimodal simulation environments, Policy optimization, Top-tier research publications, Large-scale RL experiments, PyTorch, Deep learning frameworks
Preferred skills
PhD in Machine Learning/NLP/Computer Vision, Experience with online RL settings, Track record of SOTA performance
Technologies
PyTorch, Diffusion models, Autoregressive models, Multimodal frameworks
Responsibilities
Conduct research on RL algorithms for multimodal models including diffusion and autoregressive approaches. Design and build scalable RL infrastructure for distributed training. Develop and refine reward modeling strategies to ensure stability and mitigate reward hacking. Create and curate multimodal simulation environments and datasets. Design and conduct rigorous benchmarking and evaluation protocols. Analyze and optimize policy performance across modalities. Investigate next-generation RL paradigms for environment feedback learning. Publish research findings in top-tier conferences (ICML, NeurIPS, ICLR, CVPR, ICCV, ECCV).