• Led and completed jdk migration from jdk8 to jdk17 for 30+ critical e-trading micro-services to solve the openjdk8 end-of-life issue, restored and upgraded component functionalities that support the Scotiabank’s internal mechanism of pricing automation, auto-hedging, and auto-negotiation, all components were successfully upgraded before project deadline• Performed 20+ component releases and emergency fixes, actively collaborated with multiple cross-functional teams to ensure project… Show more • Led and completed jdk migration from jdk8 to jdk17 for 30+ critical e-trading micro-services to solve the openjdk8 end-of-life issue, restored and upgraded component functionalities that support the Scotiabank’s internal mechanism of pricing automation, auto-hedging, and auto-negotiation, all components were successfully upgraded before project deadline• Performed 20+ component releases and emergency fixes, actively collaborated with multiple cross-functional teams to ensure project delivery, reduced project completion time by 20% using agile methodologies Show less

• Supervised by Professor Parham Aarabi• Built three backend APIs using Express.js to support user requests (user creation, account info retrieval, internal/external money transfer), created databases using MongoDB to store user info• Tested rigorously with 500+ parallel processes consisting of 10,000+ requests to ensure no data race nor thread conflicts, user info is also hashed with Bcrypt to ensure security even on data leaks

Built front-end applications using React.js and Human Factor principles to accelerate the internally data collection process, collected 2k+ data samples within 1 month, shortened the proposed time frame by 80%• Developed a robust back-end infrastructure using Node.js, Express.js and mongoDB, for the purpose of building a pipeline to feed more than 5k+ data points into the cybersecurity Machine Learning initiative for training purposes, ensured seamless integration with the front-end… Show more Built front-end applications using React.js and Human Factor principles to accelerate the internally data collection process, collected 2k+ data samples within 1 month, shortened the proposed time frame by 80%• Developed a robust back-end infrastructure using Node.js, Express.js and mongoDB, for the purpose of building a pipeline to feed more than 5k+ data points into the cybersecurity Machine Learning initiative for training purposes, ensured seamless integration with the front-end application and improved overall liability of the platform Show less

Robotics and Automation Laboratory (RAL) | under Prof. Andrew Goldenberg • Enabled voice recognition algorithm for robot to identify voice instruction from users, by structuring an audio library and fine-tuning keyword parameters using PyAudio and PocketSphinx, achieving high accuracy of 89% even with slight background noises• Co-developed real-time object detection algorithm for robot to locate various drinks in fridge, by structuring CNN with YOLOv4 algorithm to train on a collective… Show more Robotics and Automation Laboratory (RAL) | under Prof. Andrew Goldenberg • Enabled voice recognition algorithm for robot to identify voice instruction from users, by structuring an audio library and fine-tuning keyword parameters using PyAudio and PocketSphinx, achieving high accuracy of 89% even with slight background noises• Co-developed real-time object detection algorithm for robot to locate various drinks in fridge, by structuring CNN with YOLOv4 algorithm to train on a collective image dataset of 200k+ images, improving accuracy from 54% to 87% Show less

- Team obtained 1st-place finish in the ROBOMASTER Competition 2019