Profiles at the Edge: How Read.cv Personalizes Domains Without Custom Infrastructure
Read.cv, a professional networking platform that lets users create more personalized work profiles, relies on Vercel's Edge Functions to deliver personalized content globally with minimal latency. The service’s core feature—allowing users to attach their own custom domains to their profiles—posed a significant technical challenge that the team solved by moving away from per-domain deploys.
Previously, the engineering team considered creating separate deployments for every domain, an approach they quickly deemed unsustainable. At the same time, they were juggling infrastructure tweaks with higher-priority work like feature development and UX improvements. Migrating to Vercel enabled an incremental transition from Create React App to Next.js, offering two key gains:
- Better performance: Edge Functions serve personalized content from locations close to users, which reduces latency and keeps response times consistent worldwide.
- Increased flexibility: Read.cv can choose the right rendering method for each page—serving some directly from the edge while pre-generating others as static assets—all within a single Next.js and Vercel stack.
A crucial enabler was Vercel's support for wildcard domains, automatic SSL certificate generation, and DNS management. This combination allowed Read.cv to build a multi-tenant platform where users can point their own domains at their profiles without requiring custom infrastructure. The result is that personalization at the edge and manageable multitenancy also improved site performance and Time to First Byte (TTFB).
“We have a global customer base and it’s important to us that our speed is unbiased regardless of location. When we deploy with Vercel, users around the world are instantly served our latest site at an edge near them. With our previous provider, visitors were often being served from an edge in the U.S. even if they were in Asia or Europe.” — Mehdi Mulani, Software Engineer, Read.cv
