Research

Ideas tested in the real world.

Applied work in behavioral AI, privacy-preserving systems, and secure cloud infrastructure — developed through the Air Force Research Laboratory's HBCU/MI Summer Research Program.

Behavioral AI Security

UNMASK

"What if harmful intent only becomes visible over time?"

Behavioral middleware for evaluating concealed intent across multi-turn LLM interactions. UNMASK treats conversational behavior as structured, explainable data rather than a black-box, single-message classification.

Role: Author & Presenter Program: AFRL HBCU/MI Summer Research Program

Python Behavioral AI LLM Evaluation Explainability

Case study → Poster (PDF) → Abstract (PDF) →

Poster: Approved for Public Release · AFRL-2026-3137. Abstract: Approved for Public Release · 03 August 2026 · AFRL-2026-3479.

Secure Cloud-Native Systems

WHISPER

"How do you minimize the lifetime of both compute and credentials?"

WHISPER is a team framework for securely querying public data sources. My individual contribution was the ephemeral compute and identity layer underneath it: short-lived, scoped AWS compute that never held more access — or existed longer — than a given job required.

Role: Co-Author & Presenter Contribution: Ephemeral compute & identity layer Program: AFRL HBCU/MI Summer Research Program

AWS Lambda EC2 IAM Boto3 FastAPI

Case study → Poster (PDF) →

Poster: Approved for Public Release · 09 July 2026 · AFRL-2026-3153.

Program

AFRL HBCU/MI Summer Research Program

Both projects on this site were developed through the Air Force Research Laboratory's HBCU/MI Summer Research Program, culminating in a research symposium in August 2026.

View project case studies View experience