AROMA
Mixed-initiative AI assistance for non-visual cooking from videos.
Read the paperAccessibility Mode
High-contrast, large-type, low-motion layout with direct navigation for non-visual and low-vision readers.
Hi!👋👋👋 I'm Ning Zheng (宁政), currently an Applied Scientist at Amazon 
, P13N, building generative recommendation systems and intent understanding algorithms for Amazon homepage. I obtained my Ph.D. in Computer Science and Human-Computer Interaction (HCI) at the
University of Notre Dame 
with
Prof. Toby Jia-Jun Li.
My Ph.D. research focused on building multimodal human-AI collaboration systems for augmenting user cognitive capabilities.
Previously worked at Adobe 
and Microsoft 
as a research scientist intern.
Research Prototypes
Selected projects during my Ph.D.
Agent PbD
Agentic workflow generation from user demonstrations on web browsers.
Read the paper
PEANUT
Human-AI collaboration for multimodal video annotation.
Read the paper
SPICA
Interactive audio descriptions for blind and low-vision viewers.
Read the paper
MIMOSA
Human-AI co-creation of spatial audio effects for video.
Read the paperPublications
Peer-reviewed papers from my research.
AgentPbD: Interactive Agentic Workflow Generation from User Demonstration on Web Browsers
VL/HCC 2025
Developer Behaviors in Validating and Repairing LLM-Generated Code Using IDE and Eye Tracking
VL/HCC 2024
Insights into Natural Language Database Query Errors: From Attention Misalignment to User Handling Strategies
TiiS (IUI 2023 Special Issue)
An Empirical Study of Model Errors & User Error Discovery and Repair Strategies in Natural Language Database Queries
IUI 2023