How do folks throttle/rate-limit requests to their Realm GraphQL APIs?

Hey folks,

I’m exposing a Realm to the internet with the GraphQL API and was wondering how folks here rate-limit/throttle, or otherwise manage incoming requests to discourage abuse.

What I’m looking for here is the ability to throttle requests like AWS’s API Gateway does:

Are there any easy solutions here? Do I have to roll something myself? Should I put something in front of the API, or is it more worth my time trying to protect my work-intensive custom resolvers at the function level?