Designing and migrating applications to the Cloudflare edge, with a clear read on when it cuts cost and improves performance, and when it does not pay off.
Schedule a callData transfer, CDN, and WAF weigh on the bill. I move the edge to Cloudflare where it cuts cost without losing control.
Need per-customer isolation, SSL for SaaS, and predictable per-tenant cost at the edge.
Fast response in any region, with caching and rendering at the edge close to the user.
Semantic search and inference running close to the user with Vectorize and Workers AI, without a dedicated model server.
Workers, D1, R2, KV, Vectorize, Workers AI, and even an MCP server at the edge. What I recommend is what I operate every day.
I moved a SaaS platform's edge from AWS to Cloudflare and tracked the cost dropping month over month, toward ~85%, without pausing the product.
AWS and Cloudflare have different strengths. I choose by measuring cost and latency, case by case.
Architecture for operations with millions of visits, where edge cost and latency show up directly on the bill.
“Moving to the edge cuts cost and improves experience when you know what to move and what to leave in place.”Let's talk
When this service is relevant, the cases below show how technical judgment turns into execution and operational impact.
Case Study
Case study on cost optimization for a SaaS platform running on AWS, with strategic use of Cloudflare, per-tenant cost tracking, and a cumulative reduction close to 85%, all without pausing development.
Case Study
The engineering case behind this site: Astro on Cloudflare Workers, with D1, R2, KV, Vectorize, Workers AI, and an MCP server at the edge. Honest decisions and trade-offs.