Marco Lanfranchi
4th year Data Science student at SFU. I've done work in database management, software development, and data science.
Experience
Work
Database Engineer Intern, Samsung R&D Canada
Developed and deployed a platform that automated database account management across PostgreSQL, MySQL, Redshift, and MongoDB databases. Implemented all app functionality and introduced automated account lifecycles with password rotations and account expirations which eliminated 25%+ of DBA tickets for the company.
Data Analyst Intern, Nettwerk Music Group
Applied statistical analysis and machine learning techniques to streaming and social media data for 100s of artists under an independent label. Developed dashboards for geospatial audience streaming analytics, fraudulent stream detection, and pipelines that transformed raw streaming data into reports and visualizations.
Education
BSc in Data Science, Simon Fraser University
Volunteering
Volunteer Jr. Data Scientist, Industrio AI
Worked with a small team of data scientists and developers to build full-stack applications for fuel cell engineering clients, contributing front-end features and interactive visualizations using Python, Streamlit, Plotly, TypeScript, and Vue.js.
Research Associate, Dr. Matt Lowe, UBC School of Economics
Collaborated as an undergraduate research associate collecting data for Dr. Matt Lowe’s research project: 'Do Virtue Signals Signal Virtue?'.
Projects
iammusic-template: Web app that lets users generate custom versions of a popular artist’s album cover. At its peak, it drew over 200k visitors in a single month and has accumulated nearly 500k submissions stored in its NoSQL cloud database. The app spread widely across Instagram memes and still receives ~1k daily visitors.
digit-recognition-from-scratch: An interactive app for real-time handwritten digit classification, powered by a custom K-nearest neighbors implementation in Python.
aita-predictor: A machine learning model that classifies r/AmItheA-hole Reddit posts using an ensemble of classifiers built on vector embeddings and large-scale PySpark text processing. Presented with a Streamlit demo UI for interactive exploration.
