aiqbench takeAI Products

Uber's approach is a practical example of how enterprises are trying to make AI agents safe to deploy at scale. Separating tool registration (the control plane) from runtime execution (the proxy) mirrors long-standing API gateway design, which suggests much of agent governance is a familiar distributed-systems problem rather than a wholly new one. For AI readers, the notable part is the operational detail: automated discovery of existing APIs, LLM-generated tool descriptions, and access controls and redaction layered on top, without requiring changes to the underlying services.

Sources coveringInfoQ, Databricks, NVIDIA, Latent Space

Uber Builds MCP Gateway to Govern Access to 800+ Servers and 5,000+ AI Agent Tools

InfoQNovember 2, 20263 min read

Uber has built an MCP Gateway to standardize how AI agents discover and invoke internal services across more than 800 MCP servers and 5,000 tools. The gateway separates tool registration from runtime execution and includes automated API discovery, access controls, sensitive data redaction, incremental tool discovery through Omni MCP, and response filtering. By Leela Kumili

Uber has built an MCP Gateway to standardize how AI agents discover and invoke internal services, addressing fragmented integrations, duplicated infrastructure, and inconsistent security practices as agent adoption expands across the company. The platform supports more than 800 MCP servers and 5,000 tools, providing…

Excerpt shown under the publisher's RSS terms · 44 words · full text and figures are on the publisher's site

Read at InfoQ ↗Ask about this article