Lists / People
Jim Fan
NVIDIA Director of Robotics; created Voyager (open-ended LLM agent) and coined the "Physical Turing Test". Co-leads GEAR Lab + Project GR00T on embodied / physical AI.
Linxi "Jim" Fan is Director of Robotics and a Distinguished Research Scientist at NVIDIA, where he co-leads the GEAR Lab (Generalist Embodied Agent Research) and Project GR00T for humanoid robots. A Stanford PhD under Fei-Fei Li and OpenAI's first intern, he created Voyager — the first LLM-powered lifelong-learning agent in Minecraft — and coined the "Physical Turing Test" as the frontier benchmark for embodied AI.
39/ 100
Authority scoreEmerging
1lists#102peak1primary1category
Frameworks
Attributed to Jim Fan
Mental models, principles, and operating frameworks extracted from sources where Jim is the credited author or speaker.
6frameworks
010203040506
Code as Action
Let the agent write code as its action space — then save what works to a skill library it authors itself.
Innovation·from Jim Fan on Nvidia's Embodied AI Lab and Jensen Huang's Prediction that All Robots will be Autonomous
Data Is the Bottleneck, Not the Architecture
Robotics doesn't need a new model — it needs a three-source data engine: internet-scale + simulation + real-robot.
Strategy·from Jim Fan on Nvidia's Embodied AI Lab and Jensen Huang's Prediction that All Robots will be Autonomous
LLM-Authored Reward Functions
Stop hand-engineering reward functions — have an LLM write them in the simulator's API and iterate.
Innovation·from Jim Fan on Nvidia's Embodied AI Lab and Jensen Huang's Prediction that All Robots will be Autonomous
Specialized Generalists
Build the generalist first, then distill it down — a specialized generalist beats a native specialist almost every time.
Strategy·from Jim Fan on Nvidia's Embodied AI Lab and Jensen Huang's Prediction that All Robots will be Autonomous
The 10,001st World
Train across 10,000 randomized simulations and reality becomes just the 10,001st — sim-to-real by distribution, not fine-tuning.
Innovation·from Jim Fan on Nvidia's Embodied AI Lab and Jensen Huang's Prediction that All Robots will be Autonomous
The Foundation Agent
One model that generalizes across three axes — the skills it can do, the bodies it can control, and the realities it can master.
Innovation·from Jim Fan on Nvidia's Embodied AI Lab and Jensen Huang's Prediction that All Robots will be Autonomous
Domains
Framework distribution
6frameworks
Innovation4 frameworks · 66.7%
Strategy2 frameworks · 0%
Bibliography
Sources by Jim Fan
1source
Appears alongside
Top neighbors of Jim Fan
8people
01
Aidan Gomez
Cohere co-founder; transformer-builder commentary
1shared list
02Ajeya Cotra
biological-anchors timelines report
1shared list
03Alec Radford
GPT-1/2/CLIP communications
1shared list
04Aleksander Madry
adversarial-robustness framework
1shared list
05Alex Smola
"Dive into Deep Learning" book
1shared list
06Ali Rahimi
"Machine learning is alchemy" NeurIPS test-of-time talk
1shared list
07Allen Downey
"Think Bayes", "Think Stats"
1shared list
08Anca Dragan
assistive-AI / human-robot interaction lectures
1shared list
Last updated
29 May 2026