Profile picture

Patrick Yuan

Hey, nice to meet you! I'm Patrick, a Computer Engineering student at the University of Waterloo.

If you take interest in my work, feel free to reach out to me! I may be busy, but I'm never too busy to make new connections.

About Me

I'm currently in my 2B term of Computer Engineering at the University of Waterloo.

I'm working as an Undergradudate Research Assistant (URA) at the Vision and Image Processing Lab under the supervision of Dr. Mehrnaz Fani.

I'm also a Software Lead at UW Reality Labs, where we're currently building a VR-controlled humanoid robot in collaboration with WATonomous!

When I'm not coding, you can find me listening or playing music, cooking up a storm, or thrifting for clothes!

Projects

3D Spatial Awareness System with Audio Feedback

A 3D spatial audio navigation system for the visually impaired. The system uses a Raspberry Pi as the controller, and YOLOv11 for object detection and DepthAnythingV2 for monocular depth estimation. The system maps detected objects to 3D spatial audio sound cues, creating an immersive and intuitive user experience.

3D Spatial Awareness System with Audio Feedback

Python icon Python PyTorch icon PyTorch Raspberry Pi icon Raspberry Pi QNX icon QNX C++ icon C++ OpenCV icon OpenCV
Computer Vision Audio Processing
Sheet Music Transcriber

A web app that transcribes sheet music from audio. It uses Music.ai's API to transcribe the audio and a custom midi generation algorithmto generate sheet music.

Sheet Music Transcriber

Python icon Python JavaScript icon JavaScript Flask icon Flask React icon React Music.ai icon Music.ai OpenAI icon OpenAI TailwindCSS icon TailwindCSS
Audio Processing
Spotify Listens Timeline

A web app that visualizes user's Spotify listening history in an interactive timeline format.

Spotify Listens Timeline

Python icon Python JavaScript icon JavaScript Flask icon Flask React icon React MongoDB icon MongoDB AWS Lambda icon AWS Lambda TailwindCSS icon TailwindCSS
API Integration
ML Sign Language Translator

A real-time sign language translator that uses a custom dataset of 3000+ images to recognize 30+ static symbols and dynamic gestures.

ML Sign Language Translator

Python icon Python OpenCV icon OpenCV Scikit-Learn icon Scikit-Learn
MediaPipe Machine Learning Computer Vision

No projects found matching your selected tags.

Experience

Present
Vision and Image Processing (VIP) Lab logo

Vision and Image Processing (VIP) Lab

Research Assistant

calendar May 2025 - Present
location Waterloo, ON
  • bullet point Incoming May 2025
UW Reality Labs logo

UW Reality Labs

Software Lead

calendar May 2025 - Present
location Waterloo, ON
  • bullet point Incoming May 2025
Python icon Python PyTorch icon PyTorch C++ icon C++ Unity icon Unity Meta SDK icon Meta SDK
Deep Learning Neural Networks VR Project Management
2025
Ford Motor Company logo

Ford Motor Company

Software Pipelines Engineer

calendar January 2025 - April 2025
location Ottawa, ON
  • bullet point Migrated hardware-in-the-loop (HIL) tests to a virtualized environment using QEMU, QNX, and ARM Cortex-A processors, reducing test execution time by 80% and eliminating dependency on physical automotive hardware
  • bullet point Strategically integrated automated static analysis testing and virtualized HIL environments into the CI/CD pipeline, resulting in a 60% reduction in pre-release defects and accelerating time-to-market by 25%
  • bullet point Developed and implemented automated tests for internal static code analysis libraries, including Polyspace, Watchdog, MATLAB, and Ceedling, increasing code coverage from 67% to 95% and improving software reliability
Python icon Python C++ icon C++ QEMU icon QEMU QNX icon QNX GitHub Actions icon GitHub Actions
ARM CI/CD Jira Pipelines
UW Reality Labs logo

UW Reality Labs

Machine Learning Engineer

calendar September 2024 - April 2025
location Waterloo, ON
  • bullet point Developed and trained deep learning models for hand gesture recognition with over 98% accuracy, enabling seamless integration of advanced real-time gesture controls into Unity projects beyond Meta SDK limitations
  • bullet point Designed custom neural network architectures for single and dual-hand gesture recognition, incorporating spatial relationships like inter-hand distance and relative rotation, achieving 97% accuracy in classifying dual-hand gestures
Python icon Python PyTorch icon PyTorch Unity icon Unity Meta SDK icon Meta SDK
Deep Learning Neural Networks VR
2024
Ford Motor Company logo

Ford Motor Company

Software Engineer

calendar May 2024 - August 2024
location Ottawa, ON
  • bullet point Engineered a responsive React web app to automate vehicle software release creation, slashing developer time on these tasks by 85% and significantly increasing the CI/CD workflow efficiency across vehicle platforms
  • bullet point Developed Python scripts to automate Jira ticket creation for vehicle software test builds, reducing processing time by 80% and accelerating the release cycle by 30% for faster time-to-market for critical software updates
Python icon Python React icon React TypeScript icon TypeScript
Jira CI/CD
2023
FIRST Robotics logo

FIRST Robotics

Computer Vision Engineer

calendar August 2022 - August 2023
location Ottawa, ON
  • bullet point Developed an advanced OpenCV robot vision system, applying pose estimation to improve depth perception accuracy by 80% and enabling real-time tracking of 8 unique objects in dynamic competition environments
  • bullet point Optimized the vision system to process 30 frames per second on embedded hardware, increasing the robot's overall execution accuracy by 60% and scoring capabilities by 75% during competitions
Python icon Python OpenCV icon OpenCV
Computer Vision Embedded Systems Robotics Image Processing

Contact

Don't hesitate to reach out to me through email or LinkedIn! I'm looking forward to hearing from you!

Gengar