➢ First-hire at Kanaria (smart health toilet startup)➢ Web scrapped training data and built auto-scrape pipeline for continualdata collection➢ Managed contractors to develop AWS image processing backend➢ Wrote mobile app V2 specification documentation➢ Contributed to pitch deck development and pitching investors

https://www.beondeck.com/longevity-biotech/

ReAL (Research, Analytics, and Learning) Team➢ Accepted first-author paper to NeurIPS 2021 Machine Learning for Public Health Workshop on predicting migraine early from Fitbit data with deep learning➢ Worked with Python, PyTorch, and tsai

➢ Accepted first-author paper to Bioinformatics (Application Notes)➢ Co-built TMQuery (tmquery.gsk.com) to provide a public resource to quickly query for structurally similar proteins

Health and Human Performance Lab (https://www.healthandhumanperformancelab.com/)Our lab is unlocking the future of precision mental health. With one of the world's most comprehensive datasets of collegiate experience combining Fitbit, smartphone use, and experience sampling data, we are uncovering novel relationships between student mental health and behavior and building tools to predict and improve mental health.I led a data science team with undergrads from Stats/ML, Computer Science, Computational Biology, and Artificial Intelligence. We additionally collaborate with teams from the University of Washington, the University of Pittsburgh, and the University of Notre Dame. Work included:➢ Publication writing➢ Mentoring (ML, feature extraction, data cleaning)➢ Grant writing➢ Data collection➢ Hiring (grew team from one to eight students)➢ Interfacing with Fitbit API➢ Creating visualizations of longitudinal sleep data➢ Developing strategies to deal with noise and missing data➢ Extracting sleep features and running regression analyses➢ Creating ML model for video-based human sleep/wake classificationGrants Awarded:➢ Small Undergraduate Research Grant ($1,000 | Fall 2019)➢ Small Undergraduate Research Grant ($500 | Spring 2020)➢ Presentation Award (Summer 2020)

Ambrosio Lab | McGowan Institute of Regenerative MedicineWorked on a project to build a computational aging biomarker from scRNA-Seq data My work included:➢ Data wrangled massive scRNA-Seq datasets to build protein interaction networks➢ Interfaced with gene ontology database APIs➢ Built Python/R scripts to computationally characterize protein interaction networks➢ Worked with SLURM and the University of Pittsburgh's High Throughput Computing cluster to run scripts

Kavraki LabContributed to the Kavraki's Lab novel docking tool DINC (Docking INCrementally)My work included:➢ Benchmarked in-house protein docking tool with XSEDE supercomputer➢ Utilized Python and Javascript to add protein docking visualization features

Kumar LabAnalyzed and modeled data from the International Mouse Phenotype Consortium (IMPC) database to determine the importance of various metadata parameters and environmental variablesMy work included:➢ Interfaced with International Mouse Phenotyping Consortium (IMPC) APIs➢ Data wrangled massive IMPC dataset with AWK, Python, and Excel➢ Developed Python/R data pipeline to model environmental variance on genotype-phenotype associations from IMPC data