Evaluate before optimizing
I treat validation, calibration, and generalization as design decisions—not checks added after a model is built.
Computer Science · University of Toronto
I build and study AI systems that connect technical rigor with human needs.
I’m Qixuan, a Computer Science Specialist working across applied machine learning, human-centered visualization, adaptive media, and reliable software engineering.
- Based in
- Toronto, Canada
- Education
- U of T CS · 2027
- Experience
- Research · Veeva SDET
How I work
Three habits connect my research and engineering practice.
The common thread is careful evaluation: understanding what a system does, why it behaves that way, and whether the result is useful to people.
Build for human response
My visualization and adaptive-media work connects technical behavior to perception, interaction, and user-study evidence.
Engineer for reliability
Testing experience keeps my research practice grounded in reproducibility, careful debugging, and maintainable implementation.
Selected work
From model comparison to interactive systems.
A few projects that show how I move from a question to implementation, evaluation, and reflection.
Synesthetic Music Visualization
A TouchDesigner-based music visualization system mapping audio features to visual deformation for emotional-regulation exploration.
Box Office vs Streaming Services Visualization
An interactive D3.js dashboard analyzing box-office and streaming-service trends through coordinated web views.
Food Item Classifier
A classifier project comparing decision trees, KNN, and neural networks with feature engineering and generalization analysis.
Experience
Research thinking, strengthened by production discipline.
My engineering and research experiences reinforce the same practice: isolate the problem, test assumptions, and communicate the evidence clearly.
Jul. 2025 - Jun. 2026
Software Engineer in Test Intern, RIM Core
Veeva Systems
End-to-end automation and debugging for Vault RIM Core workflows, with attention to regression reliability and maintainable test logic.
Winter 2026
Research Manuscript Collaboration
Adaptive Music Visualization for Emotional Regulation
Manuscript research and writing connecting affective computing, cross-modal mapping, multimodal sensing, and offline reinforcement learning.
Start a conversation
Interested in thoughtful AI systems, visualization, or reliable software?
I’m always glad to compare notes on research ideas, potential collaborations, and technically grounded opportunities.