Projects
BrotherDB
Graph database of 500+ fraternity members and alumni using Neo4j and Python, enabling networking and recommendations.
Semantic Segmentation for Satellite Imagery
Self-Supervised model (U-Net + Barlow Twins) achieving 0.77 IoU for satellite imagery segmentation tasks.
JoyScore
HackTCNJ winner: Facial sentiment analysis using Google VisionAI API & OpenCV with an interactive Voila web app.
Experience
- Cleveland Clinic Akron General – AI/ML Research Assistant (Aug 2023–Present): Co-authored 4 manuscripts, created sequential NN for neonatal AKI mortality (96.5% accuracy).
- Novo Nordisk – Data Analyst Intern (Jun–Aug 2025): Integrated GPT-4o forecasting tool; analyzed 400M-row Wegovy claims data with SHAP and Dataiku workflows.
- The Data Mine, Purdue – Project Manager & Research Assistant (Aug 2022–May 2025): Led Elevance Health clustering project, ViaSat SSL research, Merck Dash application.
- Merck – Analytical R&D Intern (Jun–Aug 2024): Reduced LOQ by 10x for HPV assay; improved Python automation runtime by 250%.
Selected Publications
“Predicting Mortality Risk in Neonatal Patients with AKI with an Artificial Neural Network Algorithm” – ASN Kidney Week 2024 (co-author).
Education
Purdue University – B.S. in Computer Science & Data Science (Aug 2022 – May 2026)
Relevant Courses: ML & Data Mining, Data Structures, DBMS, AI, Statistical Theory, Linear Algebra.
Extracurriculars: Tech Committee Head @ Alpha Kappa Psi, Former Board Member @ futureofus, Quizbowl Team, Math Tutor.
Technical Skills
Languages: Python, R, Java, C, C++, SQL, NoSQL, Cypher, Git, HTML, CSS
Libraries & Tools: Dataiku, Snowflake, PySpark, Docker, Django, MySQL, Pandas, Polars, PyTorch, TensorFlow, Keras, Dash, NumPy, Matplotlib, Plotly, Tableau, Seaborn, SQLite
Contact
Email: arnavvyas25@gmail.com
LinkedIn: linkedin.com/in/arnav-vyas