Compute is becoming more heterogeneous, and programming it currently means juggling various languages and libraries in one project (e.g. Python, PyTorch, Triton, and CUDA ). The goal of Mojo is to provide a single system language for an heterogeneous era. This talk focuses on two foundations of Mojo: parametric metaprogramming and an ownership model built on origins, and the open language design problems that remain to be solved.

Chris Lattner is a co-founder and the CEO of Modular, which is building an innovative new developer platform for AI and heterogeneous compute. Modular develops MAX, an AI deployment system that accelerates GenAI and PyTorch inference, as well as the Mojo:fire: language, which extends Python into high performance and accelerator programming domains. He has also co-founded the LLVM Compiler infrastructure project, the Clang C++ compiler, the Swift programming language, the MLIR compiler infrastructure, the CIRCT project, and has contributed to many other commercial and open source projects at Apple, Tesla, Google and SiFive.