Source: https://internal-agents.com/agents/uber-coding-agent # Uber — Internal coding agent (unnamed) Uber runs an internal coding agent that writes complete code changes with no human authoring, and engineers review and approve them. Its CTO reports that engineers now delegate tasks to the agent rather than accept its suggestions, and that its share of all code changes has grown from under 1%. - Approach type: Task agent - Deployment stage: Deployed - Autonomy: Drafts reviewed - Evidence strength: Secondary only - Status: Internal - First reported year: 2026 - Work: Coding - Invocation: Unknown - Entry reviewed: 2026-09-15 ## Where people stay involved - **coding request → complete code change** — Unknown · Level unknown ## Overview ### Summary Uber runs an internal coding agent that writes complete code changes with no human authoring, and engineers review and approve them. Its CTO reports that engineers now delegate tasks to the agent rather than accept its suggestions, and that its share of all code changes has grown from under 1%. Fact · Reported · Medium confidence · `uber-coding-agent--summary` Confidence reason: Only limited public evidence supports this claim. Evidence: - Supports · [1] [Uber's CTO on AI coding agents (Business Insider)](https://www.businessinsider.com/uber-cto-ai-coding-agentic-software-engineers-2026-3) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/uber-coding-agent-source-1/content.md) ## Supervision evidence ### Operating model assessment Unclassified for coding request → complete code change; human attention boundary: unknown. Inference · Catalog judgment · Unverified confidence · `uber-coding-agent--operating-models-0` Confidence reason: The available source reports output volume but does not document where human attention returns. Qualifications: - Observation date: 2026 Evidence: - Supports · [1] [Uber's CTO on AI coding agents (Business Insider)](https://www.businessinsider.com/uber-cto-ai-coding-agentic-software-engineers-2026-3) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/uber-coding-agent-source-1/content.md) ## Implementation details ### Sandbox unknown Inference · Catalog judgment · Medium confidence · `uber-coding-agent--architecture-sandbox` Confidence reason: The preserved sources do not document an execution sandbox; unknown does not mean absent. Evidence: - Supports · [1] [Uber's CTO on AI coding agents (Business Insider)](https://www.businessinsider.com/uber-cto-ai-coding-agentic-software-engineers-2026-3) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/uber-coding-agent-source-1/content.md) ### Harness Not specified publicly Fact · Reported · Medium confidence · `uber-coding-agent--architecture-harness` Confidence reason: Only limited public evidence supports this claim. Evidence: - Supports · [1] [Uber's CTO on AI coding agents (Business Insider)](https://www.businessinsider.com/uber-cto-ai-coding-agentic-software-engineers-2026-3) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/uber-coding-agent-source-1/content.md) ## Reported metrics ### Headline claim ~1,800 complete code changes per week (~8% of changes) Metric · Reported · Medium confidence · `uber-coding-agent--headline-metric` Confidence reason: Only limited public evidence supports this claim. Qualifications: - Reported by: Uber - Scope: Code changes written entirely by the internal coding agent and human reviewed - Denominator: All Uber code changes for the reported 8% share Evidence: - Supports · [1] [Uber's CTO on AI coding agents (Business Insider)](https://www.businessinsider.com/uber-cto-ai-coding-agentic-software-engineers-2026-3) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/uber-coding-agent-source-1/content.md) · Preserved content.md, lines 24–26 ### Key observation ~1,800 complete code changes per week (~8% of changes at the time) Metric · Reported · Medium confidence · `uber-coding-agent--key-metrics-0` Confidence reason: Only limited public evidence supports this claim. Qualifications: - Reported by: Uber - Scope: Code changes written entirely by the internal coding agent and human reviewed - Denominator: All Uber code changes for the reported 8% share Evidence: - Supports · [1] [Uber's CTO on AI coding agents (Business Insider)](https://www.businessinsider.com/uber-cto-ai-coding-agentic-software-engineers-2026-3) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/uber-coding-agent-source-1/content.md) · Preserved content.md, lines 24–26 ### Key observation 95% of engineers use AI tools monthly Metric · Reported · Medium confidence · `uber-coding-agent--key-metrics-1` Confidence reason: Only limited public evidence supports this claim. Qualifications: - Reported by: Uber - Scope: Monthly use of AI tools among Uber engineers; not internal-agent-specific adoption - Denominator: Uber engineers Evidence: - Supports · [1] [Uber's CTO on AI coding agents (Business Insider)](https://www.businessinsider.com/uber-cto-ai-coding-agentic-software-engineers-2026-3) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/uber-coding-agent-source-1/content.md) · Preserved content.md, lines 18 ## Related implementations - Related implementation: [Uber — uReview](https://internal-agents.com/agents/uber-ureview) ## Sources 1. [Uber's CTO on AI coding agents (Business Insider)](https://www.businessinsider.com/uber-cto-ai-coding-agentic-software-engineers-2026-3) - News · Independent secondary · Evidence - Original URL: - Accessed: 2026-08-12 · Last verified: 2026-08-31 - Preserved copy in the repository: