I'm not a cloud-hosting fan, but comparing RDS to a single instance DB seems crazy to me. Even for a hobby project, I couldn't accept losing data since the last snapshot. If you are going to self-host PostgreSQL in production, make sure you have at least some knowledge how to setup streaming replication and have monitoring in place making sure the replication works. Ideally, use something like Patroni for automatic failover. I'm saying this a someone running fairly large self-hosted HA PostgreSQL databases in production.

RDS is not, by default, multi-instance and multi-region or fault tolerant at all - you choose all of that in your instance config. The amount of single-instance single-region zero-backup RDS setup's I've seen in the wild is honestly concerning. Do Devs think an RDS instance on it's own without explicit configuration is fault tolerant and backed-up? If you have an ec2 instance with EBS and auto-restart you have almost identical fault tolerance (yes there are some slight nuances on RDS regarding recovery following a failure).

Just found that assumption a bit dangerous. The ease with which you can set that up is easy on RDS but it's not on by default.