McGill University September 2018 - May 2022
B.Sc. in Computer Science
McGill Artificial Intelligence Society

Employment Experience

UMAknow

Software Intern, January 2020 – August 2020

• Created websites and tools on the Microsoft Sharepoint platform
• Developed better documentation reading/writing skills and more rigorous project organization skills, after working with Microsoft SPFx and TypeScript
• Developed the a platform to handle the recruiting process for the Commission Scolaire de la Baie-James using Angular. Also created a platform to handle their expenses and finances using React
• Deployed applications on Azure
• Developed better coding practices and conventions after being often helped, guided and code-reviewed by more experienced engineers and colleagues.

CAE

Software Intern, June 2019 - August 2019

• Directed and created an internal company tool called OKR-Tool alone, over the course of 2 months:
  • Web tool used by supervisors to set objectives for their teams and for team members to track and update their progress. It has a Basic Authentication login, and other features and metrics requested by the team leader
  • Created front-end using ReactJS. Created backend using NodeJS/ExpressJS. Deployed on the company private IIS server
• Wrote scripts for the company database, added features and removed bugs from company web tools using C#
• Familiarized myself with the AGILE and MVC development model and also learned working alongside experienced engineers

Projects

Artistic Animals — Web App — McGill Artificial Intelligence Society (WIP)

Github | Blog | Demo

• Generates synthetic images of animals using a Generative Adversarial Network (GAN)
• Currently attempting to implement BigGAN
• Images can then be transformed into an art style using a Convolutional Neural Network style transfer (ex: Picasso)

Largest Digit Classifier

Github

• Classifier that can detect the largest digit on an image with multiple handwritten digits
• Image data preprocessing using OpenCV to remove the noise and augment contrast, Keras to do data augmentation
• Attempted using Transfer Learning with VGG16, but ran out of time, used Convolutional Neural Network instead
• Accuracy of around 96% with more epochs and data augmentation

Proper Park — Mobile App — PolyHx 2019 (First Place)

Devpost

• Built an app in a small group as part of Hackatown. Proper Park’s goal is to decipher confusing parking signs with a single cellphone snapshot through Machine Learning
• Experimented with Phone Gap but settled on React Native in order to create the mobile interface for iOS and Android

Legends of Andor — Unity Multiplayer Game

Github

• Videogame Adaptation of the board game of the same name
• Implemented multiplayer networking using Photon
• Features of the game:
  • Turn based combat and actions
  • Multiple hero classes the players can choose from
  • Multiple enemy types
  • Usable items, gold, and inventory
  • Main questline, random events, and fog of war

SafeHouse — Web App — McGill Codejam 2019 (3rd Place IOT)

Devpost

• SafeHouse uses your home security cameras to detect if any of your doors have been left unlocked and sends you a picture and report through your Google Assistant
• Worked with Pytorch, FastAI and DialogueFlow(for Google Assistant)