Portfolio — 2026
Hello, I'm
Nicholas Conroy
Developer / Analyst / Designer
Utilizing analytics, full stack development, and creative thinking to further the goals of organizations making a difference in our world.
Projects
Real-time online versions of these projects and more additions coming soon!
Spotify Playlist Statistics
It's always fun to discover new insights about the music you enjoy, as seen by the yearly mass posting of Spotify Wrapped. I wanted to approach music insights from a different angle: instead of analyzing listening data, evaluate playlist content. The user can paste in their playlist link, and enjoy detailed and interactive analytics based on the songs within them.
Tech Stack: React Frontend, Python Flask Backend with Pandas and recharts libraries
Future Plans: Include genre analysis, possibly using a different API as Spotify’s doesn’t carry this data. Further filtering and chart types. Login system to save user data and compare with others (can opt out of being compared with)
20 Questions Game
An entertaining mini-game where users can think of an animal and see if the guessing system can correctly identify it within the question limit! If the animal is not identified, it can be suggested as an addition to the database, or the user will be informed that the animal already exists and was simply not guessed.
Tech Stack: React Frontend, Python Flask Backend with CSV for data storage
Future Plans: Use SQL database instead of CSV for animal data storage. Open up game to have more categories, and even a create-your-own game setup where the user provides the topic, questions, and a list of answers.
Shape Recognition Game
The development of young children is so important, and the variety of educational content they have plays a direct role in this. This project uses CNN-based image recognition to allow the user – ideally a small child – to draw shapes in response to a prompt and have their drawings evaluated. With each correct answer, their score goes up! There is even an option to display a happy, neutral, or slightly sad emoticon on their face dependent on their progression. The model currently accepts 3 shapes, with plans to have a more advanced model with additional shapes in the future.
Tech Stack: Python Tensorflow Convolutional Neural Network model, OpenCV image/video processing and display
Future Plans: Increase the number of detectable shapes in the model, improve UI (legend for keyboard inputs and a cleaner look). Add high scores and/or rankings, and levels of progression from simpler to more complex shapes. Allow the model to determine that the submitted drawing does not fit any possible shape if the confidence is too low (instead of always guessing a shape).