Skip to contentEmotional Storytelling in AI Filmmaking
Investigating craftsmanship in generative AI filmmaking through Research through Design and autobiographical filmmaking.
Research
I study how AI changes creative practice, with an emphasis on human agency, craftsmanship, and expression. My work spans human–AI interaction, multi-agent systems, and robot learning.
Cornell University · B.S. Computer Science, 2026 · Contact
Generative AI & craftsmanship
Cornell HCI · Advisor: Qian Yang
AI for 3D Problem Solving: Machine–Artifact Co-Design
Designing and piloting strand-based co-design tasks to study how AI supports problem–solution co-evolution as design spaces and evaluation criteria emerge.
Multi-agent systems
Cornell HCI · Advisor: Qian Yang
Upstanders’ Practicum: Interactive Multi-Agent Simulation
Developed TRUMAN AGENTS and analyzed data from a 34-participant study examining attention shifts toward active intervention in cyberbullying.
Qian Yang, Jessie Jia, Elaine Tsai, Amy Li, Nader Akoury, and Natalie N. Bazarova. “Attention: What Prevents Young Adults from Speaking Up Against Cyberbullying in an LLM-Powered Social Media Simulation.” 2026.
TrumanBench: Multi-Agent LLMs for End-User Programming
Developed a benchmark and automated evaluation pipeline for comparing LLM systems on real-world code modification tasks.
Qian Yang, J.D. Zamfirescu-Pereira, Jessie Jia, and Asad Nabi. “TrumanBench: Profiling LLMs’ Ability to Help Non-Programmers Modify a Real-World Code Base.” HEAL @ CHI, 2025.
Robot learning
Cornell Robotics · Advisor: Kuan Fang
Correspondence-Oriented Imitation Learning
Built real-to-simulation infrastructure in ManiSkill using DROID demonstrations and point-cloud assets to evaluate visuomotor policies conditioned on 3D correspondences.
Yunhao Cao, Zubin Bhaumik, Jessie Jia, Xingyi He, and Kuan Fang. “Correspondence-Oriented Imitation Learning: Flexible Visuomotor Control with 3D Conditioning.” Preprint, 2025.