Projects

Selected projects that explore distributed architecture, production-oriented backend services, data systems, and applied machine learning.

01

Distributed systems · Team project

Yet Another URL Shortener

A URL platform designed to scale beyond a single service while supporting accounts, history, and popular-domain analytics.

What we built

A React client backed by Go API microservices, deployed on AWS behind Kong and multiple load balancers.

Engineering decisions

Applied the AKF Scale Cube and sharded MongoDB clusters to separate responsibilities and distribute load.

  • Go
  • AWS
  • MongoDB
  • Kong
  • Docker
  • React
02

Backend systems · NLP

Text Summarizer Chrome Extension

A browser-based workflow for condensing long-form web content without leaving the page.

What I built

A Chrome extension with a cloud-hosted summarization service, initially in Python and later rewritten in Go.

Engineering decisions

Compared TextRank and Pointer-Generator approaches, separated the browser client from the service, and used Docker with Google App Engine for deployment.

  • Go
  • Python
  • JavaScript
  • GCP
  • Docker
  • TextRank
View source on GitHub
The extension summarizing content directly in the browser.
03

Applied machine learning · Graduate project

Emotional State Recognition

A system for recognizing emotional states using both facial expressions and body gestures.

What we built

A team-developed, containerized deep learning application. I focused on training, testing, deployment, and model serving.

Engineering decisions

Used TensorFlow Serving behind Flask and Docker to keep model inference separate from the application interface.

  • Python
  • TensorFlow
  • Keras
  • Flask
  • Docker
  • OpenCV
04

Data science

Data Science Competitions

A collection of recommendation, classification, clustering, and text-mining problems evaluated on final leaderboards.

Results

Placed 1st in book recommendation, 2nd in traffic-image classification, 3rd in news clustering, and 5th in medical-text classification.

Approach

Worked through data analysis, feature engineering, model selection, validation, and metric-driven iteration in Python.

  • Python
  • scikit-learn
  • SciPy
  • pandas
  • NumPy
View source on GitHub

Archive

Earlier projects

Smaller projects that mark earlier stages of my engineering path, preserved here without making them the focus of the portfolio.

Utopia Alexa Skill

A serverless Alexa skill built with Python, AWS Lambda, automated tests, and continuous integration.

Freesound CLI Scraper

A tested Python and Selenium tool that automated downloads of audio files and associated metadata.

Save Sister Script

A Python and Selenium utility that monitored medical-exam availability and sent an email when an earlier appointment opened.