Ingmar Kanitscheider is an OpenAI researcher best known as a co-author of "Emergent Tool Use From Multi-Agent Autocurricula," the 2019 hide-and-seek study showing agents inventing forts, ramps, and other tool strategies through self-play. Paper contribution notes credit him with obtaining the first tool-use results and co-creating the final environment with Bowen Baker.12
At a glance
- Current role
- Researcher
- Company
- OpenAI
- Fields
- Reinforcement learning, Agents, Robotics
- Sources cited
- 2
- Profile reviewed
Accomplishments
Hide-and-seek and emergent tool use
With Bowen Baker and colleagues, Kanitscheider studied multi-agent competition in a physics hide-and-seek environment where standard RL at scale produced six emergent strategy phases, including fort building and ramp use. OpenAI's blog and ICLR presentation list him among the primary authors.12
Timeline
Where they work
- OpenAI 102 people in the wiki
Related people
- Bob McGrewAlso at OpenAI
- Jim FanAlso at OpenAI
- Jeff CluneAlso at OpenAI
- Noam BrownAlso at OpenAI
- Peter WelinderAlso at OpenAI
- Kevin BlackRobotics · Agents
- Jakob FoersterReinforcement learning · Agents
- Siddharth NarayananReinforcement learning · Agents
- Deng TaihuaRobotics · Agents
- Tuomas HaarnojaReinforcement learning · Robotics
Suggest a correction or removal
Profiles cover public professional work only, and each statement cites a source. If something is out of date, missing or wrong, tell us and link a source. People can also ask for their profile to be corrected or removed.
References
2 sources · profile reviewed Oct 24, 2026 · data updated Oct 24, 2026