Projects

My projects in NLP, data science and backend development can be found on GitHub.

My favourite ones are listed below.

db-hub-fastapi Link to heading

db-hub-fastapi is a repo to experiment with building a FastAPI layer on top of modern NoSQL databases. I experiment with conventional keyword-based text stores (Elasticsearch, Meilisearch), graph-based stores (Neo4j) and vector-based stores (Qdrant, Weaviate and LanceDB). The hope is to extend this work in the future with more varieties of databases as well as APIs that expose existing data in new ways to the end user (for e.g., via GraphQL).

Gender Gap Tracker Link to heading

The Gender Gap Tracker is an automated system used to study gender bias in Canadian mainstream news media. The project was founded by a non-profit, Informed Opinions, whose goals are to amplify the voices of gender-diverse and nonbinary people in the media. I had the opportunity to work with a huge dataset of ~1.5 million news articles, scraped daily from 12 Canadian news organizations’ websites that were stored on MongoDB, and analyzed via modern NLP tools like spaCy. This work resulted in multiple peer-reviewed publications, which can be found here.