AI Is Eating SaaS: Building an IP Geolocation API in Two Hours in Rust (vpetersson.com)

In a striking demonstration of AI's impact on Software as a Service (SaaS), Screenly successfully created a self-hosted IP geolocation API in under two hours using Cursor with Claude Opus 4.5 and Rust. This new solution addresses common pain points associated with existing services like ipgeolocation.io, including API rate limits, recurring costs, and reliance on third-party uptime. The result is a lightweight, single binary of just 17MB that operates offline and provides instantaneous geolocation responses without external dependencies. The significance of this development lies in how it exemplifies the rapidly evolving landscape of software tools powered by AI. By embedding essential datasets—like the MaxMind GeoLite2-City database for IP lookups and timezone boundary polygons into the binary—the new API achieves lightning-fast response times in sub-milliseconds and remains highly efficient in memory usage. This not only minimizes operational costs but also enhances security through a reduced attack surface. As AI coding assistants demonstrate the ability to replicate entire products in a fraction of the time, the traditional value proposition of hosted services is challenged, prompting industry stakeholders to rethink their strategies in an increasingly competitive environment.