About Me

I’m an AI/ML engineer, applied researcher, and PhD candidate at the University of Maryland, Baltimore County, where my work focuses on multimodal learning and trustworthy human-robot interaction. My research explores how large language models interpret grounded instructions — integrating text, visual context, and sensor signals to help robots understand what humans mean, not just what they say.

I currently serve as a Lead AI/ML Engineer at Booz Allen Hamilton, where I lead AI/ML training, tool development, and platform enablement, and design LLM-enabled and RAG-style workflows for tagging, semantic retrieval, and knowledge discovery. Before this I built the team’s ontology-driven tagging pipeline end to end, and prior to that spent a year as a Senior Agile Engineer and Release Manager on a $200M program, where I cut average days-to-completion from 167 to 36 and blocked-release time from 250 days to zero.

Before graduate school, I built and shipped several real-world systems across mobile development, web engineering, and data-driven products. I’ve always enjoyed projects where clean engineering, thoughtful design, and tangible impact come together.

Today, my focus is split across three areas: developing data-efficient multimodal models for robotics and autonomous systems; building AI-driven productivity tools, including TimeSense; and designing reliable ML pipelines that move ideas from prototype to production. I enjoy solving ambiguous problems, turning research insights into practical solutions, and creating tools that make people more capable.