Yiyang You

What's This?

Since starting my journey at UofT, I have actively sought out professional experiences that align with my interests in computer science, statistics, and problem-solving. This page highlights roles and internships that have contributed to my growth as a developer and researcher. Each experience taught me something unique, whether it was collaborating with a team, learning new technologies, or solving challenging problems. If you'd like to know more about any of these roles or have feedback, feel free to email me at kevin.you@mail.utoronto.ca.

---

Information Technology Intern @ Alpha Education

• Led development & maintenance of the Asia Pacific Peace Museum website by implementing interactive features for the Home and Exhibitions pages using user-centric design principles, resulting in a 50% increase in website views
• Engineered two interactive applications for museum visitors using HTML, CSS, JavaScript, Leaflet.js, and Mapbox API, incorporating information from 100+ historical events, which successfully engaged 20% of visitors post-deployment

Research Assistant @ University of Toronto AI Physics & Safety Lab

• Integrated state-of-the-art TextGrad and ADAS techniques to optimize and design agentic systems for automated code generation and simulation workflows
• Enhanced algorithms for code generation, addressing convergence and over-optimization issues to improve reliability and accuracy
• Executed physics simulations using the Rebound Python package to study solar system stability, managing parameters and computational resources effectively
• Utilized the HumanEval dataset to train and evaluate AI models, ensuring high functional correctness in code synthesis tasks
• Leveraged inspect_ai and Microsoft Autogen frameworks to create multi-agent systems for simulation planning, code generation, execution, and analysis

Research Assistant with Professor Chi-Ghun Lee

• Collaborated with the FlightX research project team to explore machine learning models predicting plane arrival times by analyzing variables such as wind speed, wind direction, weather conditions, and aircraft specifications
• Conducted literature reviews and organized aviation data for the FlightX initiative by sourcing, cleaning, and structuring datasets, contributing to improved model accuracy and actionable insights
• Transitioned to Nestlé research collaboration project by taking on key roles in data preparation, analysis, and process improvements during in-office assignments, directly supporting the operational objectives of the organization
• Collaborated with Nestlé North York department's data analysts under supervision of Professor Chi-Ghun Lee to develop four predictive models for demand forecasting, promotional spend optimization, and revenue growth
• Developed a Gradient Boosting model to predict product-level demand curves with accuracy (80.4% R-squared on test data), and a 3-Layer Perceptron model for product-group level demand prediction, achieving a 72.3% R-squared on test data
• Identified 200 constraints based on business rules by analyzing pricing structures and addressing discrepancies between 'Case' and 'Each' units, resolving inconsistencies in promotional features and improving data reliability for optimization use