---
schema_version: "newruntime-agent-readable-v0.2"
type: "post"
stable_id: "post:uber-agentic-adoption-claims"
slug: "uber-agentic-adoption-claims"
title: "Uber's Agentic Adoption Claims Need Two Kinds of Evidence"
description: "An Uber employee reports 99% AI-tool adoption and agent attribution for over 70% of pull requests, while official engineering evidence confirms broad AI review at production scale."
retrieval_nugget: "An Uber employee reports 99% AI-tool adoption and agent attribution for over 70% of pull requests, while official engineering evidence confirms broad AI review at production scale. An Uber employee reports that 99% of the company's engineers use AI tools, more than 70% of pull requests are attributed to local or cloud agents, and engineers have created more than 2,500."
status: "published"
published_at: "2026-07-24"
updated_at: "2026-07-24"
record_date: "2026-07-24"
date_kind: "published_at"
topics: ["uber","ai-adoption","coding-agents"]
source_urls: ["https://x.com/praveenTweets/status/2074605343439810922","https://www.linkedin.com/posts/pneppalli_agentic-ai-adoption-is-on-fire-at-uber-and-activity-7480367291851833344-6Lhm","https://www.uber.com/blog/ureview/","https://eng.uber.com/"]
visuals: [{"id":"uber-evidence-quality-loop","kind":"editorial-diagram","role":"hero","src":"https://newruntime.com/images/posts/uber-evidence-quality-loop.webp","alt":"Employee-reported adoption claims remain separate from official Uber Engineering evidence and a multi-stage uReview quality and feedback pipeline.","caption":"Adoption claims and production-system evidence answer different questions; useful reporting keeps their provenance and definitions separate.","credit":"New Runtime synthesis from employee-reported claims and Uber Engineering's uReview report","source_url":"https://www.uber.com/blog/ureview/","generated_with":"gemini-3.1-flash-image","width":1600,"height":900,"legend":[{"label":"Reported usage","description":"An employee post reports that 99% of Uber engineers use AI tools."},{"label":"Reported attribution","description":"The same post reports that more than 70% of pull requests are attributed to local or cloud agents."},{"label":"Reported skills","description":"The employee source reports that engineers created more than 2,500 agent skills."},{"label":"Official review coverage","description":"Uber Engineering says uReview analyzes over 90% of roughly 65,000 weekly diffs."},{"label":"Official feedback","description":"Uber reports 75% of interacted-with comments marked useful and over 65% of posted comments addressed."},{"label":"Measurement contract","description":"Definitions, time windows, attribution rules, quality effects, and review load must accompany adoption percentages."}]}]
routes: {"html":"https://newruntime.com/posts/uber-agentic-adoption-claims/","markdown":"https://newruntime.com/posts/uber-agentic-adoption-claims.md","json":"https://newruntime.com/posts/uber-agentic-adoption-claims.json"}
source_format: "markdown"
---

# Uber's Agentic Adoption Claims Need Two Kinds of Evidence

## Retrieval answer

An Uber employee reports 99% AI-tool adoption and agent attribution for over 70% of pull requests, while official engineering evidence confirms broad AI review at production scale. An Uber employee reports that 99% of the company's engineers use AI tools, more than 70% of pull requests are attributed to local or cloud agents, and engineers have created more than 2,500.

An Uber employee reports that 99% of the company's engineers use AI tools, more than 70% of pull requests are attributed to local or cloud agents, and engineers have created more than 2,500 agent skills.

These are strong claims, but they are not yet an audited company-wide measurement in the public sources reviewed here. They should be attributed to the employee post, not restated as independently verified Uber statistics.

Official Uber Engineering material does confirm a substantial production footprint. Uber's uReview system analyzes over 90% of roughly 65,000 weekly diffs, uses specialized generation and grading stages, and continuously evaluates model combinations. That is direct evidence of AI embedded in the engineering control plane.

## Why does the distinction matter?

"Uses an AI tool" can mean anything from occasional completion to sustained agentic work. "Attributed to an agent" depends on how authorship is recorded when a human, local agent, cloud agent, and reviewer all touch the same change.

Useful adoption reporting should publish definitions alongside percentages:

- active use over what time window;
- tool access versus meaningful use;
- generated code versus agent-attributed pull requests;
- accepted output versus output later rewritten;
- quality, incident, and review-load effects.

## New Runtime Read

The headline is not that humans disappeared from Uber engineering. The stronger signal is that AI has moved into shared infrastructure: review coverage, model evaluation, feedback capture, filtering, and staged rollout.

For our project, adoption should be measured the same way. Count verified outcomes—sources resolved, drafts accepted, incidents avoided, deadlines met—not agent activity alone. A 99% usage number is interesting; an auditable quality loop is operationally valuable.