Hi, I'm Ryan Divigalpitiya
Welcome to my portfolio
Large Language Models' Ability to Capture Meaning in Unique Vocabularies
2023
Social media users tend to utilize memes and slang which constantly evolve as internet culture shifts over time. For my MSc research, I’ve been tasked to develop techniques that explore how well LLMs are capturing the meaning of such jargon words found in these unique vocabularies.
Large Language Model-Powered Workout App for iPhone
2023
In my spare time, I’m developing a weight-lifting workout app that will use a locally-deployed Large Language Model (LLM) that will generate workout plans for specific muscle groups and convert routines written down in the iOS Notes app into workout plans managed within the app.
Startup: Nexus Labs
2018 — 2020
For my second startup, I built a convolutional neural network paired with thermal sensors that were then used to detect room occupancy in a smart-home. My early prototype achieved 99% accuracy in discriminating between false-positive and true-positive signals.
Startup: Perpetually Innovative Homes
2016 — 2018
I co-founded a company that set out to build highly energy-efficient smart homes. Our builds utilize advanced technology to deliver more affordable, environmentally-sustainable living. The company is growing and expanding to build in various locations across Ontario. I left in 2018 to start Nexus Labs.
CS 9860 Advanced Machine Learning: Final Project
December 2020
For my CS 9860 Advanced Machine Learning Course, I proposed and built Convolutional Neural Networks paired with inexpensive thermal sensors to recognize room occupancy in a smart-home. My model achieved ~99% accuracy in discriminating between false-positive and true-positive signals.
CS 2212 Introduction to Software Engineering: Final Project
April 2016
In collaboration with Fitbit©, my final project in CS 2212 Introduction to Software Engineering was to build a Fitbit© desktop application for Windows that could link to a user’s Fitbit© device. I was project manager and my team won 1st place for the best app in our class.
CS 9637 Data Science: Final Project
December 2020
For my CS 9637 Data Science Course, I proposed, built and tuned a random-forest multi-classifier that can predict the probability of someone falling into a class of obesity given that they maintain their current lifestyle choices and habits. My model achieved 87% accuracy in predicting obesity.
CS 3307 Software Engineering: Final Project
April 2018
In collaboration with the Schulich Medical School and the Department of Computer Science at Western University, my final project in CS 3307 was to build the next version of a data visualization application used by the medical school at Western. I was project manager for my team and developed back-end features.
CS 9552 Human-Computer Interaction: Final Project
April 2019
For my CS 9552 Human-Computer Interaction course, I was the project manager for my team's final project. I proposed and took lead on designing an app that could help students identify careers that interest them and help them plan their academic path to best prepare for those careers.
MapDesk UI / UX Proposal
2018
On the side, I took several freelancing jobs in software development, UI / UX design, and graphic design. A startup called Normap (Toronto, Ontario) hired me as a freelancer to create UI designs of their future dashboard web app.
Cylix Front-End Design & Development
2017
A company called Cylix (London, England) hired me as a freelancer to (1) create a refreshed UI design for their aging desktop and mobile interface and (2) develop their new front-end of their desktop app using Qt (in C++) based on my UI designs.