← Home

Research

My research focuses on the usability of differential privacy (DP) libraries and tooling: how API design, documentation, and developer workflows affect correct, practical use of DP in real systems.

What is Differential Privacy?

Differential privacy (DP) is a rigorous mathematical framework that limits what can be learned about any individual from released aggregate data by adding calibrated noise to computations. It is widely used in academic research and production systems to balance utility and privacy. See Wikipedia for a formal overview.

Key open-source projects

• OpenDP — community-driven open-source DP tools and working groups.
• Google Differential Privacy — DP building blocks and frameworks.
• IBM diffprivlib — Python library for experimenting with DP in ML.

Research goals

I study developer-facing challenges: API ergonomics, testing & auditing DP code, common pitfalls in parameter choices (ε selection), and tooling to detect privacy regressions. The aim is to make DP tools easier to adopt while preserving formal guarantees.

Join the discussion on Discord: discord.gg/bbJQppWjr