> Markdown version of [/jobs/ext/4622420-research-scientist-robot-learning-rl](https://www.wearedevelopers.com/jobs/ext/4622420-research-scientist-robot-learning-rl). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md).

---

# Research Scientist - Robot Learning & RL

- **Company:** Acceler8
- **Location:** San Francisco, CA, United States
- **Contract:** Permanent contract
- **Skills:** Artificial Intelligence, Catalyst (Software), Python (Programming Language), Machine Learning, Reinforcement Learning, Pytorch, Embodied AI, Information Technology
- **Published:** November 7, 2026
- **Apply:** https://www.juju.com/job/28_25197_04f0927d827d1e0f62da558688112db1

## About the Role

* PhD in robotics, ML, computer science or equivalent original research experience
 * Strong publications at venues such as NeurIPS, ICLR, ICML, CoRL, RSS, ICRA or IROS
 * Deep reinforcement learning knowledge and hands-on policy development
 * Strong Python and PyTorch skills
 * Experience with simulation, embodied learning, robotic control or interactive environments
 * Enthusiasm for exploratory research and full-time onsite work in San Francisco
 * Imitation learning, world models, sim-to-real or human-AI cooperation experience is especially valuable.

## Description

I'm working with a frontier AI research lab building low-latency models that help AI understand, act and cooperate with people in real time.

Starting with games and simulation, they're researching how agents learn, coordinate and adapt alongside humans-and how these capabilities could transfer into robotics.

Supporters include Supercell, General Catalyst and South Park Commons, alongside participation in NVIDIA's Inception Program.

This is an opportunity to own original research and help shape an early direction spanning reinforcement learning, embodied AI and human-robot collaboration.

You'll work on:

 * Developing policies using imitation learning, offline RL and online RL
 * Building simulation environments and evaluations for control, planning and cooperation
 * Researching world models, hierarchical policies and adaptation across tasks
 * Exploring transfer from games and simulation into robotics
 * Connecting learned policies with perception, memory and conversational behavior