Hi,
I have some experience with those kinds of services and it's quite time consuming. Nowadays you can use amazing services like AWS and to at least build a proof of concept shouldn't be that hard. If the traffic is relatively low you can even self host a solution just to validate the idea. The tech stack is not the hardest part - it's content. A problem involving a slow tech stack is a good problem, that means there's demand. A great tech stack that's not used is the worst scenario.
A suggestion might even to be to buy an off the shelf solution that works OK for low volume traffic just to get started and then level up once you have traction. The key is still content - can you engage users? Will they return? Will you retain them? Do you produce content yourself? Do you license/purchase other's content?
The point is, don't focus too much on tech at this point and try to validate your idea of content, because that will be your differentiating factor in the end. Good luck!