Person profile
Senior Research Scientist at NVIDIA.
Tools Jim Fan has publicly discussed or demonstrated using. Curated by Magpie's editorial team from podcasts, posts, and public stacks.
Jim Fan is a Senior Research Scientist at NVIDIA focused on robotics and foundation models. He's behind projects like Voyager and Eureka and posts near-daily on X about frontier-model research, robotics, and the path to embodied agents.
Fan benchmarks ChatGPT alongside other frontier multimodal models in his research commentary on X. He covers OpenAI product launches and uses ChatGPT as the canonical reference point in his analysis of LLM and VLM capability boundaries, especially as it relates to agent benchmarks like CaP-Bench that evaluate 12 frontier LLMs/VLMs across 8 evaluation tiers.
Fan discusses Claude alongside ChatGPT in his X commentary on frontier-model capability boundaries. He treats Claude as one of the canonical models tested in his agent-benchmarking research, particularly in the agentic-harness work that systematically varies model choice, API abstraction level and visual grounding methods.
Fan posts Midjourney visualizations alongside research commentary on X, particularly when illustrating embodied-AI scenarios or futurology threads about the path to physical AGI. He uses Midjourney as the visual layer for ideas his text alone can't convey, treating it as a working creative tool in his research narrative.
Last researched May 10, 2026
Editorial note
These tools have been mentioned by the people featured in podcasts, interviews, posts, and other public media. Source links and citations are included wherever possible. Magpie AI has no direct relationship with the people on these pages. Spot a missing tool or a wrong attribution? Suggest a correction →