Designing secure and scalable distributed applications in most programming languages requires an unreasonable amount of vigilance. With my advisor Andrew Myers, I study language-based approaches to security and distributed computing that enable developers to express these properties directly, yielding programs that are secure and performant by construction.

Current Projects

• Fabric
  A language and system for scalable and secure distributed computation on persistent objects.
• Jif
  An extension of Java with static information flow control.
• Polyglot
  An extensible compiler framework.

Publications

• Sharing mobile code securely with information flow control (Oakland'12)

  A new architecture for secure mobile code, supporting secure use, publication, and sharing of code across trust domains.

• Automatic partitioning of database applications (VLDB '12)

  For consistently good performance, Pyxis automatically migrates computation between the client and the database server.