Just my 2 cents, I don't/wouldn't use a service like this on anything other then an MVP or non-critical internal tool.
I say this because if you base your entire business on a 3rd party service like Parse was or like Firebase etc, your entire business is at their mercy and whim. If they decide to remove a feature you rely on you are potentially stuck trying to rapidly recreate it. If the service gets bought and the new acquirer decides you are a competitor or that they wanted the team but don't care about the service, everything can disappear. Of course, I am not advocating you never use/depend on a 3rd party, cause we all have too, but just that having your entire critical core infrastructure not being something you own is not smart long term. FWIW, Firebase as it matures within the Google ecosystem would be one that I'd be more likely to use in production. Technically my current solution does have Firebase integrated into Production but we are removing it as I disagree with storing any critical production data in a "database service" where it isn't something I can migrate easily. e.g. we use postgres in AWS as the primary datastore, I could use postgres in GCP/Azure or in my own datacenter and not change any code, but I can't do that with Firebase.
I do think services like Firebase etc are great for MVP's or for some proof of concept work, I just don't like them for live production services. I also have used them for non-critical internal tools or tools which have a short lifespan etc.