Data Science Intern
Engaged in weekly stand-up meetings, fostering project advancement and idea exchange, aligning with agile software development principlesIteratively refined code driven by team feedback on GitHub pull requestsProject: Extracting food distributor names from product descriptionso Operated on dataset of 7 million rows and 4 columnso Tokenized distributor names --> performed Jaro-Winkler fuzzy string matching on tokens --> reviewed results to refine & tune algorithmo Reduced false positive cases by detecting natural keywords within food categories (e.g. ignored common words like Steak)o Consistently scored removal accuracy of 95%+ on random samplesProject: Customer recommendation algorithm for food products to switch to/fromo Analyzed food purchase behavior to detect long-term switcheso Spearheaded creation of system to assign flavor score ratings from customers to food products they purchasedo Trained machine learning model to perform matrix factorization (collaborative filtering) on flavor score vs customer-product matrixo Documented code and decisions in detail for future maintenance by team