Portrait of Todd Mytkowicz

Todd Mytkowicz

Principle Researcher


I think hard about abstractions that help programmers easily express complex problems yet are sufficiently constrained such that we can build runtimes to efficiently implement those abstractions.  Broadly speaking, I build abstractions and runtimes for (i) probabilistic programming and approximate computing and (ii) parallel programming.