Summary
Vidushi Dadu is a Senior Systems Research Engineer at Google Research with a PhD in Computer Architecture from UCLA and eight years of experience designing programmable hardware accelerators for datacenter-scale workloads. Her work focuses on hardware-software codesign and general-purpose spatial accelerator mechanisms that exploit data-dependence patterns across ML, graph, database, and signal-processing domains. She has driven research and applied engineering at Google, Intel, and Microsoft—building trace-driven tools for spatial architectures, characterizing SQL and Spark hotspots, and prototyping runtime/compiler strategies for Edge TPU dynamic shapes. Vidushi’s PhD contributions include TaskStream and execution-model primitives that expose task scheduling and inter-task communication to hardware, enabling accelerator-like efficiency for task-parallel workloads. Based in the United States, she blends deep academic insight with production-focused system design, often surfacing cross-cutting execution behaviors rather than application-specific features.
8 years of coding experience
Bachelor’s Degree, Electronics and Communications Engineering, Bachelor’s Degree, Electronics and Communications Engineering at Indian Institute of Technology , Roorkee
Doctor of Philosophy - PhD, Computer Architecture, Doctor of Philosophy - PhD, Computer Architecture at University of California, Los Angeles