RL AI Research Intern

ResearchRemote (US/Singapore Preferred)Internship — 12 to 16 weeks

Embed directly with our research team to tackle open problems in reinforcement learning for AI agents.

About the Role

As an RL Research Intern, you’ll embed directly with our research team to tackle open problems in reinforcement learning for AI agents. This is a hands-on research role where you’ll design experiments, implement algorithms, and contribute meaningfully to Pokee’s core technology—with mentorship from researchers who have led RL platforms at Meta AI and studied under pioneers like Rich Sutton.

What You'll Do

Investigate and prototype novel RL approaches for context selection, reward shaping, or policy optimization in agent workflows
Design and execute experiments, producing clear analyses and actionable insights
Implement and benchmark RL algorithms against internal and public baselines
Collaborate with senior researchers and engineers in a fast-paced startup environment
Document findings and present results to the broader team; strong work may lead to publication opportunities

What We're Looking For

Required

Currently pursuing a PhD or Master’s degree in Reinforcement Learning, Machine Learning, Computer Science, or a related field
Solid understanding of RL fundamentals (MDPs, policy gradients, Q-learning, actor-critic methods)
Strong programming skills in Python with experience in PyTorch or similar frameworks
Ability to independently design experiments and iterate quickly on ideas
Excellent written and verbal communication skills

Bonus Points

Publications or preprints at top ML venues
Experience with LLM fine-tuning, RLHF, or AI agent frameworks
Familiarity with distributed training or large-scale experiment infrastructure
Prior industry research internship experience

Who You Are

You want to join a small, elite team solving one of the hardest problems in AI—building agents that actually work in the real world. You’ll have direct impact on the product, access to cutting-edge research, and the opportunity to shape the future of enterprise AI from the ground up.