Research
I currently work on the foundations of secure distributed computation in realistic settings. My guiding question is what it takes for cryptographic protocols to remain secure and efficient once they leave the whiteboard. That means running over incomplete or asynchronous networks, against adversaries that may corrupt a majority of the participants, and alongside many other protocols at the same time. I approach this question from several connected directions: secure multiparty computation, fault-tolerant distributed algorithms, composable security, and generic black-box compilers. I combine impossibility results and lower bounds with new constructions, aiming for protocols that are not only provably secure but also practical to deploy.
Secure Multiparty Computation
Secure multiparty computation (MPC) lets mutually distrusting parties jointly compute on their private inputs without revealing anything beyond the result. I study the foundations of MPC protocols in realistic settings: incomplete communication networks, asynchronous communication, and a dishonest majority of corrupted parties. I consider both computational and information-theoretic security. The central questions are what can be achieved in these settings, and at what cost.
Related papers
- Randomized Early-Stopping Byzantine Agreement with Applications to Round-Efficient MPC — Manuscript
- On the Round Complexity of Dishonest-Majority MPC — TCC 2026
- Universally Composable Almost-Everywhere Secure Computation — JoC
- Is It Even Possible? On the Parallel Composition of Asynchronous MPC Protocols — TCC 2025
Distributed Systems
Fault-tolerant distributed algorithms are important in their own right, and they are also at the heart of many cryptographic protocols. I study Byzantine agreement, broadcast, and distributed randomness generation (e.g., common coins), which is often the key ingredient for fast agreement. So far my focus has been on round complexity, which is key to making them practical. This involves both proving lower bounds and designing protocols that improve the upper bounds. Our results here have also had direct applications to round-efficient MPC.
Related papers
- Randomized Early-Stopping Byzantine Agreement with Applications to Round-Efficient MPC — Manuscript
- On the Round Complexity of Dishonest-Majority MPC — TCC 2026
- Concurrent Asynchronous Byzantine Agreement in Expected-Constant Rounds, Revisited — TCC 2023
Composable Security
The ultimate goal of a cryptographic protocol is to be deployed. Once deployed, it runs in a potentially adversarial environment, alongside many other and possibly insecure protocols. It is therefore natural to analyze protocols in frameworks that guarantee secure composition, such as Universal Composability (UC). When needed, I also study and extend these frameworks so they capture our growing needs. Composable security runs as a parallel thread through all of my other projects.
Related papers
Black-Box Transformations & Compilers
Another part of my work concerns black-box transformations and compilers. These are generic methods that automatically turn an existing solution for one setting or requirement into a solution for another, in as general a way as possible. I study the limits of such transformations and the conditions under which they are possible. They are interesting in theory, and also useful in practice: they let us adapt existing solutions to new problems without opening the box.
Related papers
Blockchain & Decentralized Applications
Blockchains bring all of the above together: consensus, secure computation, and composition in an open, adversarial environment. Through a research internship at Input Output Global and industry-sponsored projects, I work on the composable treatment of blockchain protocols, smart contracts, and decentralized applications.
Ongoing work; papers in preparation.