LLM Practitioner Engagement
Mixed-methods study (surveys + interviews) examining how engineers perceive and trust LLMs across enterprise and start-up contexts. Investigates how practitioners' mental models shape real-world AI deployment decisions.
PhD Researcher & AI Engineer
Harris Yang
PhD student at the University of Toronto studying how people perceive and trust AI systems in the real world. I build LLM systems, teach analytics, and occasionally release rap albums.
01
About Me
I'm a PhD student in Information at the University of Toronto, where I study how people perceive and trust AI systems in practice. My research uses surveys and interviews to understand how engineers form mental models of LLMs — and how those models shape the decisions they make when deploying AI in production.
AI Researcher
AI trust & perception, practitioner studies, mixed-methods research
ML Engineer
Multi-agent pipelines, fine-tuning, production LLM systems
Educator
Teaching predictive analytics & AI at the University of Waterloo
Rapper
6 remix albums on Spotify and NetEase Music
02
Featured Projects
Automated Invoice Processing Pipeline
Multi-agent LLM pipeline on Azure automating financial invoice processing for enterprise clients. Includes OCR, dynamic schema determination, and human-in-the-loop checkpoints in a CI/CD framework.
LLM Applications Survey
Survey study examining how university students adopt and perceive LLMs in academic settings. Findings informed curriculum redesign at the University of Waterloo.
Financial Scenario Generation Model
LSTM-VAE model for financial time series reconstruction and stress testing. Used to generate risk scenarios for portfolio evaluation and train a reinforcement learning trading agent.
Let's Work Together
I'm always interested in hearing about new projects and opportunities. Whether you have a question or just want to say hi, feel free to reach out!
Send Me an Email