Memli is the ultimate flashcard app geared toward personalized learning. It incorporates adaptive learning and generative AI to help alleviate common pain points of flashcards. Effortlessly remember everything you need, for free! We're live so check us out in the iOS and Google Play stores.
PennCloud is a cloud-based system designed to provide web services, including: file storage, mail, and administrative functions (akin to Google Drive and Gmail). PennCloud handles user requests through a load-balanced network of frontend servers each running on the HTTP 1.1 protocol. Frontend servers match client requests using request handlers, which connect to a backend system in which data is stored and accessed through an abstraction of the distributed key-value store (akin to Google BigTable).
TopSpin is an advanced analytics application motivated by the age old question in sport, "who is the greatest of all time?" Through historical tennis data, the app highlights player performances and provides insight into how players across eras would have stacked up against one another. View players, tournaments, and matches; using machine learning, users are also able to simulate theoretical tennis competitions. TopSpin is built in full stack javascript, with a MySQL database and Python based ML-engine.
Acoustic Print is a data science application that takes a unique approach to acoustic fingerprinting. Based on audio features of a song, such as energy, tempo, and valence, to name a few, a two-part acoustic fingerprint is constructed using polar curves and visualized in 3-dimensions. Acoustic Print allows users to visually perceive similarities and differences between songs based on underlying characteristics.
A script that provides useful recommendations and suggestions for where you should travel next based on some of your past experiences. Decisions can be informed by up to 3 past travel experiences or destinations on your wish list; you can query the system for similar locations to those provided or dissimilar destinations if you're looking for a change, and it will generate up to 25 suggestions for you to consider from over 200 possibilities.
An EdTech platform that provided undergraduate students with heightened transparency into, and increased control over, their academic performance through the use of analytics
An implementation of the LeNet-5 convolutional neural network used to recognize handwritten digits from the MNIST data set. The implementation uses Tensorflow 2.0 and is based on the architecture from the original paper, with some alterations.
Implementation of a discrete-event simulation model based on a bank tellers staffing system. From this, conclusions are drawn about what this analysis might imply for a bank, or more broadly a business using cashiers or tellers (at a basic level).
I like to work on things related to machine learning (broadly defined), statistics, and math. I hope to continue creating cool things worth sharing... stay tuned ✌