This is great for any time you have a mostly read-only database (which is true for most publishing use-cases - anything you might consider a static site generator for) which you want to provide API access to with lightning-fast speed because replies will be served from a CDN edge.

The data served from the edge can be modified without touching code, just produce a new version of your SQLite database.

A small DB to search in a list of storefronts which typical retail websites have, a database of flags or promotion codes which you don't want to send to the browser are some examples I came up with.

Personally I would write a code generator to embed the data in the code but to each their own.

Cloudflare workers run at the edge along with a simple but fast KV store.

We use KV and workers to serve static data (json) to browser based code.

This could be used to filter, sort or aggregate that data before sending the client.

Perhaps for inventory / product lookup on a e-commerce site.

I use something like this for a metadata search tool serving static data. Its high performance and easy to set up which is the main use case for me.

With this, suppose you want faster response than Disk, then every serverless function can have its own replica of data & power responses off of that. It is like a cache through sqlclient. But wait, can't we do that already running in memory SQLLite? why is it a big deal now?

Nothing. Demoing novel wasm usage maybe?

Reinvent http proxies, but now with additional code and data that gets out-of-sync and is harder to debug.