Built an AI-powered MediaPipe eye-tracking system for pathologists' gaze analysis on 500+ digital pathology slides, improving model training efficiency by 33%. Conducted EDA on 9k+ gaze datapoints and integrated results into computer vision pipelines.
Developed and deployed a cross-platform iOS/Android app with LLM-powered trivia engine for South Asian Alzheimer's patients. Analyzed 2,200+ patient interactions and integrated gamified metrics, improving cognitive outcome evaluations by 22%.
Performed EDA on 30k+ daily production records and engineered ML features that improved defect detection accuracy by >98%. Built Python/Django data pipeline on AWS with React dashboard, reducing waste by 27% and boosting efficiency by 18%.
Boosted user engagement by 16% and cut page load time by 23% by building full-stack features and implementing Three.js visualizations. Analyzed client usage data to drive feature prioritization, directly acquiring 12 new clients in Indiana.
View my LinkedIn for more
Building an AI assistant to automate business tasks via Google Calendar, Slack and Customer Relationship Management. Utilizing PostgreSQL, FastAPI, NLP to manage business data, optimize workflows, generate actionable insights.
Built an ML-based anime recommender to provide recommendations for anime watchers with a 92% accuracy. Applied clustering, collaborative filtering to train the model, optimizing data pipelines with Pandas and Numpy.
Built a platform using MongoDB, Express, React, and Node.js for enhancing shopping for a Nepali home goods store. Developed REST APIs, used TypeScript for frontend/backend, implemented authentication, improved accessibility.
Trained a TensorFlow model on Google Colab, reducing model size by 2x to deploy by converting it to TFLite. Crafted a Python Script using OpenCV and NumPy to achieve real-time object detection, processing 30 FPS.
Created an ATM Simulator in Java using JFrame and Maven, handling up to 55 transactions running parallelly. Integrated MySQL with Java Database Connectivity, improving data management efficiency and privacy.
View my GitHub for more
I see science as more than formulas—it's a mindset I apply across computer science, data science, machine learning, and software engineering to connect numbers with people. My focus is simple and human-centric: building technology that improves decisions, streamlines workflows, and ensures data serves people, not the other way around. With a BA in CS from DePauw, and a BS in Data Science plus an MS in Data Engineering & Statistics at WashU (with certificates in Financial Engineering and ML), I bridge rigorous math with real-world systems.