This is a "Now page"—it details what I'm focused on right now, at this point in my life.

I'm currently a machine learning engineer at the Software Engineering Institute (SEI), a federally funded research and development center (FFRDC) affiliated with Carnegie Mellon University, think of it as similar to MIT Lincoln Labs. My work focuses on autonomous systems, with a recent emphasis on unmanned surface vessels (USVs). I wrapped up a study examining the current capabilities of autonomous USV systems to help inform senior leadership about the state of the practice. The findings were in part responsible for the Department of the Navy's reorganization to better handle autonomous systems acquisition.

In late November 2024, I was awarded $1.8 million in congressionally allocated research funding for a two-year project focused on course of action planners for autonomous systems. I'm partnering with CMU's AirLab to apply some of this work to disaster response scenarios. I built a UI with automatic speech recognition (ASR) and text-to-speech (TTS) to allow a disaster response coordinator to interact with the planners in a natural way. The back-end orchestration uses LangGraph to interpret user intent and route to the appropriate subgraph. One of the planners is an agentic system that writes PDDL (Planning Domain Definition Language), building on An End-to-End Planning Framework with Agentic LLMs and PDDL by La Malfa, Zhu, Marro, Bernardini, and Wooldridge at the University of Oxford.

I'm a lifelong learner. Last year I transferred from SEI's Arlington, VA office to Pittsburgh to have better access to campus resources. This fall, I took a course through the Robotics Institute on Human-Robot Interaction which I wrote about here. I completed the SolveIt course from Answer.ai and liked it so much that I built CellMate, a Python library for embedding LLM interactions directly into Jupyter Notebooks, designed for working locally when dealing with proprietary, confidential, or otherwise sensitive data. I'm also working on stdinfer, which uses tmux and Ollama to provide a local LLM in the terminal that can see your history and respond to your questions.