TAKATOSHI LEE

ML, AI, Software Engineering | AI for All | Computer Vision Enthusiast

I'm Taka, a second-year Computer Science student at the University of Toronto passionate about building AI tools that empower everyone—not just experts. Currently, I'm creating a peer tutoring platform to help UofT students successfully navigate the POST admissions process. Let's connect and build something impactful together!

Projects
Logify journaling app interface showing mood tracking and sentiment analysis
HACKATHON
Logify
CSS
HTML
JavaScript
LocalStorage
OpenAI API
Spotify API
Logify is a journaling app that analyzes your journal entries, extracting sentiment and emotional context. Based on the identified mood and writing patterns, it offers tailored comments and recommends Spotify songs that match emotional state. The app visualizes mood trends on an interactive calendar, helping you keep track your mental well-being over time.
DanceSync
Python
TensorFlow
OpenCV
Mediapipe
FFmpeg
NumPy
Matplotlib
Your smart dance partner. DanceSync compares two dance videos frame by frame, analyzing pose, timing, and technique. With real-time feedback, pose tracking, and detailed scoring, DanceSync helps you perfect your moves and synchronize like a pro.
Logify journaling app interface showing mood tracking and sentiment analysis
HACKATHON
Logify
CSS
HTML
JavaScript
LocalStorage
OpenAI API
Spotify API
Logify is a journaling app that analyzes your journal entries, extracting sentiment and emotional context. Based on the identified mood and writing patterns, it offers tailored comments and recommends Spotify songs that match emotional state. The app visualizes mood trends on an interactive calendar, helping you keep track your mental well-being over time.
DanceSync
Python
TensorFlow
OpenCV
Mediapipe
FFmpeg
NumPy
Matplotlib
Your smart dance partner. DanceSync compares two dance videos frame by frame, analyzing pose, timing, and technique. With real-time feedback, pose tracking, and detailed scoring, DanceSync helps you perfect your moves and synchronize like a pro.
Research
Minimum Spanning Tree clustering algorithm visualization and research results
RESEARCH PROJECT
Clustering with Minimum Spanning Trees
I compared two Minimum Spanning Tree (MST)-based construction algorithms, Kruskal's and Prim's, to the popular K-Means approach. After benchmarking both on synthetic datasets, Kruskal's performed better and was then tested against K-Means across 200+ real-world datasets using the Adjusted Asymmetric Accuracy (AAA) metric. While K-Means performed more consistently, MST-based clustering showed a clear advantage in detecting non-convex patterns and irregularly shaped clusters, which K-Means struggles with due to its assumption of spherical clusters around centroids.
Awards
  • Singapore National Olympiad Finalist
  • Top 250 on CodeBreaker.com
  • MLDA Hackathon Finalist @NTU
  • Top 150 of 5000 teams on Wharton Global High School Investment Competition
Education
International Baccalaureate® (IB) Diploma
2011 - 2024
Overseas Family School
University of Toronto
2024 - 2028
Computer Science & Information Security Specialist
Minor in Entrepreneurship & Business Science
Work Experience

AI Research Intern

DiSa Digital Safety June 2022 - July 2022

Evaluated state-of-the-art AI engines to craft customized songs based on user-defined parameters.

  • Contributed to refining AI algorithms and user experience, enhancing the overall music generation process.
  • Analyzed and benchmarked multiple AI music generation models.
  • Developed user interface prototypes for parameter customization.
  • Collaborated with a team of 5 researchers to improve algorithm accuracy.
Contact