1. Request comes in
You call the API with a URL and options: viewport size, device
profile, format, wait strategy, headers, cookies (for your own
properties) and more. We validate the payload, attach a trace ID
and route it to the render queue for your account.
2. Browser session & lifecycle
A headless Chromium instance is started in a sandbox with the
requested settings. We navigate to the URL, execute JavaScript,
load fonts and styles, and watch the network and DOM until
your chosen “ready” condition is met.
- Network idle or fixed delay
- Wait for a CSS selector to appear
- Hard timeouts to avoid stalls
3. Capture & delivery
Once the page is ready, we capture a screenshot or PDF at device
resolution, apply optional cropping, encode the output and stream
it back. You can choose to keep assets for a short time window or
to treat every render as truly ephemeral.