Professional Experience
Tech: Golang, Python, Scala, BigQuery
- Working on Community and Interest Targeting
Tech: Java, React, Typescript, GraphQL, Hive, Snowflake, Google Ad
- Led successful features, experiments, and large code refactors for ads across the site. Significantly increasing ad revenue through improved impressions, viewability, and CTR
- Improved header bidding by cleaning up several ad targeting values and integrating a cookieless approach that associates users by their email addresses across different platforms.
- Increased reporting capability by fixing and creating new live traffic data collection pipelines for analysts, resulting in increased client retention through comprehensive tableau dashboards
- Independently planned and executed monthly roll-out of Travelers Choice campaigns. Coordinating redesign efforts, bug fixes, and server maintenance across our related products / microservices / databases
Tech: Elastic Search, Jenkins, Groovy, Docker, Python, Bash, Openshift, UrbanCodeDeploy, SSL
- Proposed and led development of production server deployment pipeline for Beats and Logstash used in monitoring of jmx metrics and logs from Jira \& Jenkins, overall saving teams hours debugging failing servers
- Automated various tasks for audit requests and internal management like deletion of unused permission groups and unnecessary backups from Github tape storage saving the company approximately $3000 monthly
- Volunteered to maintain open source fork of Searxand mentor participants for an internal company innovation challenge. Searx was adopted by several teams to resolve search problem at RBC with chat bots
Tech: Tech: React, Redux, Jenkins, Selenium, Java, Scala, PostgreSQL, Jupyter, Spark, Docker
- Developed critical user-facing features (tables and notifications) requiring React/Redux and resolved several bugs before release. Providing users decreased initial load times by 90% speed and more useful feedback
- Wrote regression tests for ETL pipeline (data layer) using Luigi, Yarn UI, and SQL queries. Also refactored integration tests for application layer using Selenium+Jenkins, identifying several bugs before release
Project Experience
Plan-it was initially a project pitched to us at beginning of our Intro to Software Engineering course. The project aimed to differentiate itself from existing platforms, by catering to planning needs of users. I worked in a team of 5 students to draft user stories and gather requirements, and deliver the final product in a 2 month time frame, while following scrum/agile and adminestering it through Jira. We were invited to the bridge innovation award, for being the best team to envision the product owners idea for the Plan-it project (award pending, post covid-19)
Tech: Java, Spring, HTML, CSS, Thymeleaf, MySQL, MongoDB, Docker, Jira
- Implemented the user authentication, session handler, and user privledges using Java's security library. Passwords were hashed into a MySql database
- Implemented the reviews system, created the user page, and back end integration with mongodb
- Dockerized application so it would be consistent for our group members
Let Me Know, 2016-2017
Web Development, News Aggregation website
LetMeKnow was initially a successful project a couple of colleagues and I developed over at MHacks8 (top 10, and contenders for Microsoft Prize). Since October of 2016 myself the other team members have been making incremental changes in our spare time to the software. In November 2017 we entered LMK into the UTSC Startup Competition and received seed funding and access to great mentors/resources.
Tech: Python, Javascript, Django, Angular, Nginx, PostgresSQL, DigitalOcean
- Implemented multiprocessing across web queries fetching relevant articles from Reddit API, tweets from Twitter’s API, and images from Bing’s API reducing load time from 20 seconds per user query to 3 seconds
- Developed core news search utilities using Django for RESTful API management and PostgreSQL for database queries. Managed hosting for website with load balancing through Nginx on a AWS VPS
Quadbot is a FOSS chat bot available for multiple platforms, like: discord, and pokemonshowdown.com. It was initially created due to a lack of LaTeX support on these platforms.
Tech: Python, Asyncio, Image Hosting, Linux CLI, DigitalOcean
- Setup Travis CI for unit/integreation tests written with pytest, and deployed inside a docker container
- Created and maintained several commands to entertain users
- Calc: Integreated calculator functionality with the GNOMECalc software
- Dilbert:returns url to today's dilbert comic
- Latex: Designed system for users to compile LaTeX commands and receive a url with the result
- Markov: Generates a sentence based on room chat data using a Markov Chain model
- Periodic Parser: Generates a word that can be spelled with the symbols for periodic table elements.
- Party Parrot: Generates a random party parrot
- Putnam: Generates a random Putnam problem
- Xkcd: returns url to a randomly generated or specified xkcd comic
Mobile Waste
Wizard, 2015
Mobile Development, Waste Categorization App Demo for City of Toronto
The Mobile Waste Wizard is a mobile Android application that is based off of the City of Toronto's waste wizard website. This application is cable of correctly categorizing objects according to the standard set out by the City of Toronto's waste management team. It is also capable of voice recognition for those in a rush! Published article: here.
Tech: Android, Python, Java, XML, Google Material Design
- Parsed publicly available data, and incorporated it into search functions
- Redesigned the UI by following Google Material Design pattern
- Added linters, and some test automation.
