About
A breadth-first way into hard problems.
I build AI systems that have to work on messy, real-world text, then I try to measure whether they actually do.
Now
Heading back to UNH.
This fall I’m returning to the University of New Hampshire for a PhD on agent evaluations: scoring behavior against intent in knowledge bases, RAG, and memory systems.
Before that, I founded ResAlign AI, a career copilot for fit and preparedness. The research bug started as a master’s class project and became a CheckThat! 2025 paper.
Before
The path is not linear.
MS, Computer Science · UNH · 2022–2025
IT Help Desk · UNH · 2024–2025. Made me faster at finding the actual failure.
SAP Security Consultant · TCS · 2021–2022. Roles, authorization, SoD—the boring, load-bearing layer.
I was never a LeetCode person. More first principles, breadth over depth, and top-down thinking.