Data Scientist with full stack software development experience. I am a bit of a jack-of-all-trades, having studied pure math at the University of Waterloo, spent time at large and small companies in a variety of development and data roles, and founded my own startup-turned-open-source-project Polychart. As of Fall 2016 I am a masters student at the University of Toronto.
My proudest achievements are:
- Speaking at the Spark Summit East conference in 2016.
- Scoring 33 points on the Putnam math contest (where median score is 0 most years).
- Publishing on the Facebook data science blog.
- Writing three blog posts that made the front page of Hacker News.
- Founding a startup Polychart, taking it through accelerator program Extreme Startups, securing over $135,000 in funding, meeting awesome people for a ride that culminated in an open source project and personal growth.
Favourite tools of mine include Python, R (with RStudio, ggplot2), coffeescript, vim, git, Spark for big data, logisitc regression, and Bayesian techniques.
Relevant Past Work
Senior Data ScientistRubicon Project, Toronto, Ontario, Canada. Feb 2016 - June 2016.
Improved and extended the ensemble of algorithms used for ad targetting. Added modelling based on site and IP information leveraging GraphX, and presented the model at the Spark Summit East 2016. (Python, Spark)
Data ScientistChango / Rubicon Project, Toronto, Ontario, Canada. Sep 2014 - Feb 2016.
Built a product recommendation engine for dynamic advertisments, and a suite of front-end tools to automate new client setup. Built an A/B test framework for dynamic ads. Completed a move of a complex, ensemble of algorithms used for ad targetting to spark. Research work on identifying bots and suspicious traffic. (Python, Spark)
VP Engineering & Data ScientistRubikloud Technologies, Toronto, Ontario, Canada. Winter 2013 - Summer 2014.
Analyzed e-commerce traffic data to detect factors that increase conversion and revenue. Led a team of 3+ engineers and data scientists to build a real-time anomaly detection system. (R, Cassandra, Storm, MongoDB, Python, CoffeeScript)
Founder & CEOPolychart Inc, Kitchener/Toronto, Ontario, Canada. Fall 2011 - Winter 2013
Software Developer (Intern)ContextLogic, San Francisco, California, USA. Summer 2011
Built a pipeline on Hadoop to generate model files used for topic inference of text. Sped up topic inference while improving quality. (Python, Hadoop, C++)
Data Scientist (Intern)Facebook, Palo Alto, California, USA. Spring & Winter 2010
Studied the impact of "negative events" on user engagement. Explored a machine learning approach to photo recommendation. Built web-based data tools used by other analysts. Used Hadoop and R heavily. (R, Hadoop, Hive, PHP, C++)
Software Developer in Analytics (Intern)Tagged Inc, San Francisco, California, USA. Summer 2009
Improved the in-house analytics platform used to track metrics and perform A/B testing. Built frontend features used by millions of users world wide. (PHP, GreenPlum)
Random Things I Won
People's Choice Award at DataInsightSFData Visualization Competition, San Francisco, California, USA. Summer 2011
Hacking Education: Data Analysis TieSponsored by Donors Choose, New York, New York, USA. Summer 2011
My analysis of DonorsChoose's dataset tied first in the Data Analysis category of the "Hacking Education" competition. After this I did some pro bono work for DonorsChoose.
Velocity Venture Fund Pitch CompetitionVelocity Venture Fund, Waterloo, Ontario, Canada. Fall 2011
Won $25,000 to kick-start startup Polychart.
Joint Honours Pure & Applied Math, Bachelor of Mathematics, University of WaterlooWaterloo, Ontario, Canada. 2007 - 2012
President of the Statistics Club. Part of the Velocity residence, a community of student entrepreneurs and founders.