concurrent.cd

“Across Uber's vast codebase and data ecosystem, comprising hundreds of millions of code lines and thousands of tables, agents spend most of their turns locating information rather than generating code. To address this, we engineered the AI Context Graph: a unified network containing 24 million nodes and 80 million edges across 86 nodes and 117 edge types.”

— Uday Kiran Medisetty, Uber · Context Engineering

Quote · uber.com

Read the source on uber.com

From “Agents spend most turns locating information”, found 2026-10-16.

Take

Context lookup is the main cost of agents at company scale, and shared context is the infrastructure fix. Also from the post: "More than 70% of pull requests are attributed to local or cloud agents." Published 2026-08-27.

AI-drafted