Agentic SDLC at Uber — Uday Kiran Medisetty & Adam Huda, Uber
Uber's agentic AI investments now generate 70% of PRs and doubled engineer code output year-over-year.
“70% of our PRs now either by local or cloud agents. And all of this led to twice the number of lines of code per engineer year-over-year.”
Uber engineers shared that agentic AI now accounts for over 70% of pull requests across their thousands of engineers at 12 global sites, doubling lines of code per engineer year-over-year. They also automated 250+ migrations covering 9 million lines of code. Their approach is built on six building blocks anchored by an internal model gateway that enforces PII redaction, safety guardrails under 100ms latency, and per-user cost attribution — providing a concrete enterprise-scale blueprint for agentic SDLC adoption.