I am an AI engineer and part-time researcher working on the reliability of language models (LMs). I am fortunate enough to have received the NSF CISE Graduate Fellowship (CSGrad4US) and am applying to CS PhD programs for the Fall 2027 cycle.
Research
I interested in how we can make LMs more reliable for high-stakes, knowledge-intensive tasks (e.g., legal, policy, cybersecurity). My work tackles this from two angles:
- Better Evaluations: understanding when/if model behavior degrades when tested under new conditions.
- Better Training: extracting useful signals for training and evaluation when there are no easy ground truths.
Dayjob
I am currently an AI engineer within the R&D department at Five Strand Technologies, where I work on rapid AI prototyping for anti-money laundering, on-chain analytics, and federal compliance features.
Some things I've done in the past:
- AI Governance: Contributed to GPAI hazard taxonomies aimed at policymakers in the CEN-CENELEC JTC 21, the committee drafting technical standards for the EU AI Act.
- Federal DS: Worked on automated NLP/text analytics platforms for the U.S. Department of Defense (NSIN) and the United States Navy.
- Industry DS: Worked on customer analytics at Adobe and loan-process analytics at PNC Financial Services.
Personal
Beyond my professional and academic interests, I am an avid fan of all things gamified problem-solving: board games, escape rooms, speed puzzling, and murder mysteries. I also enjoy rucking and old-fashioned donuts.