serverless VS nodejs-api-starter

Here is the actual package used [1]. I thought it would have been similar to serverless-pg[2] with initiation outside the handler as with the node implementation. Any insights into how they achieve connection caching (similar to pooling?) on the edge?

[1]: https://github.com/neondatabase/serverless

(1) React + API monorepo VS (2) React + API as separate repositories

Nope. Just plain GraphQL.js + Knex.js (example), often typed manually since it's not where most of the time spend on when building an API server (though it can be automated via a scaffolding script if there are too many db tables). Generating TypeScript definitions off the actual db schema using knex-types (example).

Examples: - https://github.com/kriasoft/node-starter-kit - basic Node.js API starter kit - https://github.com/kriasoft/graphql-starter - monorepo template with GraphQL, React, Relay

https://github.com/kriasoft/graphql-starter - See web/relay.config.js, run: yarn relay

Your first example is the code-first development approach (example), while the latter one is schema-first, popularized by Apollo.

Source code (example): https://github.com/kriasoft/graphql-starter

Here is yet another example of a Node.js (Express) middleware that encodes/decodes JWT (OAuth) token into a session cookie: https://github.com/kriasoft/graphql-starter/blob/main/api/session.ts

Details: It's going to be a Reddit-like community but for books. Just started working on this idea last week (solo) by branching out from graphql-starter open source template. Looking for feedback and feature requests.

Yes, for example: