Annabelle Cortez

Annabelle is a systems engineer by training and an infrastructure obsessive by temperament. She holds a PhD in Electrical Engineering and Computer Science from UC Berkeley, where she developed novel approaches to fault-tolerant distributed training that are now standard practice at several major AI labs.

Before co-founding Drane Labs with James, Annabelle spent eight years at Google, first on the Borg cluster management team and later as a principal engineer on the TPU infrastructure group, where she led the architecture for their third-generation training pipeline. Her work on scalable checkpointing and recovery systems has been cited in over forty peer-reviewed publications.

Annabelle left Google in 2022 after growing frustrated with the gap between what large organizations said about responsible AI and what their infrastructure actually made possible. At Drane Labs, she's built the technical foundation from the ground up — training infrastructure, evaluation frameworks, and the internal tooling that lets a team of thirteen punch well above its weight. She's particularly proud of the company's open-source model evaluation suite, which has seen adoption at over sixty organizations worldwide.

She holds three patents in distributed systems architecture and was named to MIT Technology Review's Innovators Under 35 in 2020.