Layers runs on AWS, Cloudflare and PlanetScale. It uses Polylane to triage its alerts and fix the issues that reach its merchants' storefronts.
Connected to Polylane
The challenge
Search is where a shopper decides to buy. Layers answers every search, collection page and recommendation on its merchants' storefronts, and it promises them 99.99% monthly uptime. A slow query, a stale index or a failed sync shows up as lost sales on someone else's store.
The team ships many times a day across AWS, Cloudflare and PlanetScale, and its merchants set the traffic: a launch, a drop or a sale can multiply load in minutes.
“We promise our merchants 99.99% uptime, and a sale on their store does not wait for us to wake up. Polylane reads every alert from Better Stack and Sentry, works out which ones matter, and usually has the fix open before we have finished reading the page.”
Jake Casto Founder at Layers Why Polylane
Polylane builds a context graph of Layers' stack across AWS, Cloudflare and PlanetScale and connects it to its repositories, Sentry, Better Stack and PostHog. It learns what normal looks like for each path a search takes, raises an issue with the evidence when something drifts, investigates across code, deploys and telemetry, and opens the fix when the cause is in Layers' code.
How Layers uses it
Alerts from Better Stack and Sentry go to Polylane first, which folds the noise and raises the issues that matter in Slack and Linear. Fixes go through Layers' normal review and CI.
The results
In Layers' first month, Polylane triaged 2,220 alerts, detected 734 issues and opened 45 fixes, with a median of 12 minutes from detection to a pull request. Layers merged 34 of them (76%).