Atlas Archive
Atlas needed a public archive that could host 12,000+ catalog records without melting under their existing traffic spikes. We built a content-first site with aggressive caching and a faceted search experience.
- Web Design
- Web Development
The existing archive was a 12-year-old PHP site that buckled under traffic during the museum's quarterly free-admission weekends. Search was effectively broken; the page weight was north of 8MB.
Static-first Next.js with on-demand ISR for catalog updates, edge-search powered by a hosted search API, and an aggressive image strategy (AVIF + responsive + lazy).
Page weight down 92%. Search p95 under 200ms. Traffic spikes now invisible — the site serves 30× its previous peak with the same infra.
Mosaic Rebrand
Have a project like this in mind?
We take on a small number of engagements at a time. If this feels like your kind of work, we should talk.