We are accustomed to the power of universal computers: a single machine that can execute any computation. Scalable proof systems have the same property (theoretically): one set of constraints can enforce the computational integrity of any computation.

But what's the best way to achieve this? What are the tradeoffs between expressiveness, efficiency and code soundness? Concretely, how do domain specific languages (DSLs), Harvard and Von Neumann architectures compare?

In this episode we'll discuss these questions with members of StarkWare grappling with these questions as they build the Cairo language and VM, alongside StarkWare's Michael Riabzev (co-founder and Chief Architect), Lior Goldberg (Head of Algebra Research) and Shahar Papini (Engineer).

One STARK to Rule Them All