I am a PhD candidate at Stanford University advised by Fredrik Kjolstad. My interests include programming language design and optimization. Specifically, how do we adapt to the agentic era, where proofs are “free”, optimizations are “free”, but token consumption (and its inevitable environmental impact) is not.
Toward this end, we’ve designed Isomer, a functional programming language that decouples optimization of data layouts via rewrite rules from the source program. Because every rewrite preserves semantics, {performance engineers, matrix multiplications} can search over a space of correct implementations without spending {time, tokens} on re-verifying each candidate.
Previously, I’ve worked on the Bonsai programming language and compiler, XLA (Accelerated Linear Algebra) compiler, Calyx compiler infrastructure, CIRCT compiler for improving hardware design tools, and LLVM. Science is a team effort, and I have been fortunate to learn from and work with really good researchers, among them (in no particular order) Adrian Sampson (Cornell), AJ Root (Stanford), Clément Pit-Claudel (EPFL), and Rachit Nigam (MIT).
In my free time, I enjoy climbing, surfing, and reading. As for nomenclature, I go by Chris.
Talk on decoupling data layouts given at the Portal retreat.
Extended visit and (informal) talk given to Équipe CASH at ENS Lyon.
Bonsai: Compiling Queries to Pruned Tree Traversals won the distinguished paper award at PLDI ’26.
Attended the Programming Language Implementation Summer School (PLISS).
Visiting the SYSTEMF lab at EPFL.