I work on AI, robotics, and research systems.

My current interests are centered around building reliable embodied AI systems: robot learning, policy evaluation, simulation-to-real-world transfer, and structured experiment management for research codebases.

• Robot learning and embodied AI
• Vision-language-action systems
• Imitation learning and reinforcement learning
• Safe and reliable policy deployment
• Simulation, MVP environments, and real-world robot evaluation
• Reproducible research workflows and experiment infrastructure

I like research projects that are not only technically interesting, but also reproducible and easy to inspect.

That usually means:

• clear experiment specifications;
• well-organized training and evaluation scripts;
• explicit metrics and result files;
• traceable experiment runs;
• documentation that makes it possible for another person or AI agent to continue the work.

• research-project-guide: guidance for organizing research projects with clear docs/, code/, and runs/ responsibilities.

• Website: jialiangfan.com
• GitHub: github.com/JialiangFan