I've become increasingly aware of how much damage AI can do when it's misused, from deepfakes to AI-powered scams to cyberattacks. The same technology that makes AI so useful is what makes it dangerous in the wrong hands, and I can't stop thinking about that.
That's why I want to build in cybersecurity, AI governance, and ethics instead of studying the problem from the sidelines. I'm a computer science student at Northeastern University with an AI concentration, and my research on LLM accountability and transparency led me to start building projects like mcp-trustcheck, a scanner that flags risky MCP servers before developers connect their AI tools to them.
I also come to this as a creative. I love art, media, and design, and I think safe AI has to be something people can understand and want to use, not only something that works on paper.
Qualified for regionals 4 times and state 3 times. Placed 2nd at state and won the Moderna Award.
2021 to 2025Went to state every year. My senior capstone became a provisional patent.
2022 to 2025Taught K-12 teachers about engineering design at MIT, and talked about it on PBS NewsHour Classroom.
2022 to 2025Turned monthly trash collection data into charts on environmental quality for Charlotte, NC.
2023Northeastern AI Club (AINU)
Part-time and hybrid, on the club's creative side. I built its brand identity guide: a color palette, tone and messaging guidelines, an AI-prompt flier template, and a mock Instagram grid.
Break Through Tech at Cornell Tech
Selected for the AI Studio, a project-based extension of the fellowship. I'm building an LLM compliance auditor that checks software requirements against NASA's NPR 7150.2, using RAG and structured output verification so every verdict cites the clause behind it. I co-own the retrieval evaluation, scored on a hand-labeled set of 50 clauses.
WISE, Women Interdisciplinary Society of Entrepreneurship
Supporting a cohort-based startup accelerator that takes student founders from early idea to MVP. I develop curriculum, act as the main contact for cohort members, source speakers, coordinate events, and build deliverables through the final showcase. I lead the program with two teammates.
Break Through Tech at Cornell Tech
Selected from 4,000+ applicants for a year-long fellowship through Cornell Tech focused on responsible AI. I completed the 9-week Machine Learning Foundations course, from data prep and linear models (I built one from scratch) to neural networks, LLMs, RAG, and agentic AI, through weekly live labs.
Break Through Tech AI Studio
An LLM auditor that checks software requirements against NASA's NPR 7150.2, using retrieval and structured output verification so each verdict cites the clause behind it.
I co-own the retrieval evaluation: precision and recall on a hand-labeled set of 50 clauses.
An MCP server trust scanner
Flags risky or malicious MCP servers before developers connect their AI tools to them. The static layer is built, with six checks per tool and fuzzy matching for look-alike names. A runtime layer is planned.
A reading tracker
A full-stack app for tracking what I read, with sign-in, stats, and a customizable look.
An AI security company, in the idea stage
One package that protects people from scams and AI attacks, and helps developers and companies build safer AI.
Starting with one focused piece first.
A paper on how LLM chatbots can be secured and held accountable, drawing on frameworks from OWASP, NIST, and the EU AI Act.
Draft in progress. I'll post it here when it's out.
Ask me for the draft