First: databases
Probably SQL, as the schema is well defined.
In past projects I did a bare metal MySQL in the living room. Couldn't afford to pay for a 500gb table / 300gb ram machine in the cloud obviously. Worked well, but I could never figure out how to set up security to safely expose to the internet. I think I'll keep this as my airgapped data science machine.
Last night I tried to spool up SQL servers on google cloud compute. Did it, but omg what a time suck. Also seems quite expensive.
Tried on Azure as well. Once again, felt like a ton of work
Looked at Firebase. Super slick and easy, but NOSQL confuses me. Seems like a rigid schema is my friend here. Went looking for a similar experience but in SQL.
Tried CockroachLabs. Seemed promising, but had trouble getting connected. Probably my own ineptitude.
Finally hit PlanetScale. Super slick. Got to my first table read/write from python in 10 minutes. Seems lightweight, but can scale. I like the free tier, 10 billion row reads for free. I may run out of storage if this grows big. Luckily this project is just for me.