I'm Vansh Patel, a software engineer based in Lexington, Kentucky. I currently work at The Nuclear Company, where I build digital tools used on nuclear construction sites — hierarchical mapping interfaces, data models for field walkdowns, and automated ingestion pipelines for construction data.

I graduated from the University of Kentucky in 2025 with a Bachelor of Science in Computer Science, a Minor in Mathematics, and a Certificate in Artificial Intelligence. I made the Dean's List in Fall 2023, Spring 2024, Fall 2024, and Spring 2025.

Before joining The Nuclear Company, I built AI research tooling at the Center for Computational Sciences at the University of Kentucky — including a transcription pipeline twice as fast as the incumbent solution — and did applied geospatial ML research at the Center of Applied AI, predicting opioid overdose hotspots for RADOR-KY.

My work spans machine learning systems (forecasting, embeddings, reinforcement learning), full-stack web platforms (Next.js, TypeScript), and iOS apps built in Swift and SwiftUI. Recent projects include TOKENUINELY, an MCP server that gives coding agents semantic retrieval over large repositories, and BIG_TALK, an iOS speech-practice app with on-device transcription and fluency scoring.

I'm open to software engineering roles — particularly teams working on applied machine learning, developer tools, or products where thoughtful engineering decisions matter. If you're recruiting, the fastest way to reach me is email at vanshpatel1023@gmail.com.