CHARITH MENDIS

Charith Mendis is an Assistant Professor at the University of Illinois at Urbana-Champaign. His broad research interests include programming languages, compilers and machine learning. He envisions a future where most compiler optimizations will be auto-generated as opposed to manually written. Some of his contributions include multiple data-driven cost models for different architectures such as x86, Google TPU and automated compiler construction techniques using formal logic targeting vector and tensor hardware. The Google TPU cost model is already used in production daily to optimize fleet wide ML workloads. He is the recipient of a best paper award at ML for Systems workshop, a best student paper award at IEEE BigData and an outstanding Masters Thesis award at MIT. He received a PhD from Massachusetts Institute of Technology.