Internships

Here is a sample run of the ML service I created. You can see the various clusters represented by different colors, especially the yellow and orange ones, indicating low-healthcare quality in those regions and customer groups.

WhizAI

Leading a team of 4, I created an unsupervised Machine Learning Clustering service using 5 algorithms (DBSCAN, OPTICS, GMM, K-Means, and Hierarchical) to analyze pharmaceutical drug distribution data across various factors to identify healthcare inequities. The service is implemented as part of the company’s offerings and works to reduce healthcare disparities in underserved areas. As a result of my work, my internship was also extended from 10-weeks to 1 year.

Standard Partners Management

Looking into the SEC’s Electronic Data Gathering, Analysis, and Retrieval (EDGAR) system, I created an algorithm to automatically extract financial and product data about any company in any market. Using this information, I was able to create digital twins of companies and replicate markets and industry connections through a Neo4j database.

This is a demonstration of the data extraction algorithm for a market segment as well as its visualization using a Neo4j graph database. In choosing Tesla as the initial company, the program collected all financial information related to Tesla, as well as that of its competitors in the market. All of this data is compiled in the Neo4j database through industry connections.