Bio

My name’s Josh. My experience is in data science, analytics, programming, public policy, and marketing. You can view my up-to-date resume here.

After a hiatus, I’m back at school finishing my degree at the University of Waterloo in Science and Business. Initially to work on the provincial election campaign as a data scientist, and then to help Destination Ontario build its analytic/strategic capacity.

My academic interests have been in the computational sciences (specifically neuroscience) and biological statistics.

Projects

My Github contributions over the last year:

jtloong's Github chart

And here are some of the things I am working on or have worked on recently. You can see more them more thoroughly on my Github.

DescriptionPicture
Created a toy LSTM text generation model based on Canadian federal parliamentary debates.
Helping the University of Waterloo Data Science Club build its first website which will host club information, their blog and a knowledge repository of everything data science.
Made a tiny package that wraps around Pandas to encode continuous variables into binned binary features.
Developed git-status-size a custom git utility to track the file sizes of new repo changes and allow users to add files to their .gitignore.
Assisting a small cognitive science research team at McGill University with studying human facial recognition to better understand Dual Process Theory.
Currently building a web demo of Google’s sg2im neural network to build photorealistic images from scene graphs using React.js and Flask.
Constructed a pipeline for analyzing social media marketing performance based on image features using object-detection and colour extraction. (Closed Source)
Built a poll aggregation model for the Ontario provincial election using state-space modeling with PyStan and Plotly. (Closed Source)
As part of Code Across Toronto 2018 I was part of a small team that built an experiment in participatory democracy for Toronto�s Parkdale neighbourhood.
Contributer to Timeliny a jQuery plugin for building interactive web timelines. I mainly contributed navigation features to the library.
Created a Flask web app to allow users to extract colour palettes contained within photos.
Helped launch Chalmers Cards a guerrilla project bringing information resources to people who live in unstable housing conditions.
Used open data from the city of Toronto to create an interactive map visualizing the city�s shelter locations and their capacity.
Technology development at Democracy Kit a non-profit providing educational resources and technology for candidates running for municipal office.
Built the first Python module for computing Chow Break tests to measure structural breaks in time series data.
Using Mapbox.js and a (now-deprecated) Job Bank API, allowed users to visualize the locations of employment opportunities posted to the federal job board.
Wrote a paper analyzing government marketing data to provide insights into how the Ontario Cabinet Office can better communicate public policy.
Co-founded BykMe a now defunct startup that aimed to bring peer-to-peer bike sharing to Kitchener-Waterloo.