New capability:

The AI engineering stack we built internally — on the platform we ship

Cloudflare·Cloud·NewCapability·4/20/2026

We built our internal AI engineering stack on the same products we ship. That means 20 million requests routed through AI Gateway, 241 billion tokens processed, and inference running on Workers AI, serving more than 3,683 internal users. Here's how we did it.

Why it matters → New functionality is available in Cloudflare. May enable simpler implementations or replace external dependencies.

Who should care → Teams already using this product or evaluating it.

View original source ↗

Source payload preview

{
  "guid": {
    "#text": "4z7ZJ5ZbY3e3YwJtkg6NiD",
    "@_isPermaLink": "false"
  },
  "link": "https://blog.cloudflare.com/internal-ai-engineering-stack/",
  "title": "The AI engineering stack we built internally — on the platform we ship",
  "pubDate": "Mon, 20 Apr 2026 13:00:00 GMT",
  "category": [
    "Agents Week",
    "Agents",
    "AI",
    "Cloudflare Workers",
    "SASE",
    "MCP",
    "Developer Platform",
    "Developers",
    "Cloudflare Gateway",
    "Product News",
    "Workers AI"
  ],
  "dc:creator": [
    "Ayush Thakur",
    "Scott Roe-Meschke",
    "Rajesh Bhatia"
  ],
  "description": " We built our internal AI engineering stack on the same products we ship. That means 20 million requests routed through AI Gateway, 241 billion tokens processed, and inference running 
…