How Uber Runs 60,000 AI Agent Tasks Per Week With MCP

This Agentic AI Foundation blog post summarizes a talk by Meghana Somasundara and Rush Tehrani at the MCP Dev Summit North America 2026 about Uber's production-scale MCP deployment. Uber runs 60,000 AI agent executions per week, with over 1,500 active agents monthly and more than 90% of Uber's 5,000-plus engineers using AI tooling every month. The infrastructure is built on MCP, which the authors state "are what make AI usable at Uber." Before MCP, every agent team built bespoke integrations to Uber's 10,000-plus internal services, resulting in hundreds of non-reusable parallel integrations. The solution was a control plane consisting of the MCP Gateway and Registry, which automatically translates Uber's service interface definitions (proto and thrift files) into MCP tool descriptions using an LLM. Everything runs through code as pull requests with security scanning before deployment. Security layers include authentication on by default for sensitive data, a PII redactor service, continuous code scanning, and guardrails blocking write operations to critical services. Three consumption surfaces are supported: Uber Agent Builder (no-code platform for thousands of internal agents), Uber Agent SDK (code-first for production-grade agents like grocery assistant and customer support), and coding agents (Claude Code, Cursor, and Uber's Minions which produces roughly 1,800 code changes per week used by 95% of engineers). The roadmap includes evaluation metrics and service SLAs in the registry, a tool-search tool that finds other MCPs on demand, and a Skills Registry for reusable workflow knowledge. The authors present the MCP Gateway and Registry pattern as a reproducible model for production-grade MCP deployment at scale. 

https://aaif.io/blog/how-uber-runs-60000-ai-agent-tasks-per-week-with-mcp/

Comments

Post a Comment

Popular posts from this blog

Prompt Engineering Demands Rigorous Evaluation

In the blog post “Prompt Engineering Requires Evaluation” on the Shostack + Associates website, the author argues that treating prompts for large language models (LLMs) merely as creative artefacts is insufficient. Engineering prompts properly demands structured evaluation frameworks — what the AI community calls “evals” — to test which prompt versions work better, with which models, and under which conditions. The post highlights that simply assuming a prompt is “good enough” creates risks when LLMs are integrated into production systems (e.g., for threat modeling). It advocates for measuring prompt performance, variation effects, and tool-chain dependencies (model, context, ancillary materials).  Ultimately the message is: prompt engineering should borrow disciplined practices from software engineering (versioning, testing, benchmarking) rather than relying on informal experimentation.  https://shostack.org/blog/prompt-enignieering-requires-evaluation/
Read more

SecObserve: Simplified Vulnerability and License Management for CI/CD Pipelines

 SecObserve is an open-source tool for managing vulnerabilities and licenses in software development and cloud environments. It integrates various vulnerability scanners into CI/CD pipelines using GitLab CI templates and GitHub Actions for streamlined setup. It offers a centralized dashboard for assessing and reporting vulnerabilities, with tools for filtering, sorting, and evaluating results. SecObserve supports automation and manual assessments to focus on resolving critical issues.  https://github.com/MaibornWolff/SecObserve/tree/dev
Read more

Secure Vibe Coding Guide: Best Practices for Writing Secure Code

Ken Huang's "Secure Vibe Coding Guide" emphasizes the importance of integrating security into the software development lifecycle. The guide provides best practices for writing secure code, including input validation, proper authentication mechanisms, and secure data storage techniques. It also highlights the necessity of regular code reviews and staying updated with the latest security vulnerabilities and patches. By following these guidelines, developers can create applications that are resilient against common security threats and contribute to a safer digital environment.  https://kenhuangus.substack.com/p/secure-vibe-coding-guide
Read more