There are some things on the data ingestion side and warehousing that we've been doing on our own server and we're deciding whether to build into AWS or just forklift it later on.
Without getting into specifics, my question would be how important you think it is to optimize for AWS early on. I'm of the view that if a product is "born in the cloud" it will pay dividends down the line. I've seen things that need to be Retrofitted to the cloud later and it can be a disaster.
Keep in mind we are in the MVP phase, so naturally we are very limited on engineering resources and would like to reduce the amount of time spent on DevOps / infrastructure stuff.
Would appreciate feedback on this. Thank you HN
In our case, we host a Next.js app on Vercel, knowing if we needed to we could lift and shift everything into AWS later if necessary.
As I see it the value proposition for cloud native is: (1) scalable, (2) simple to implement. You get those two at the cost of something which is often less efficient and more expensive to run than something you run yourself.
From the viewpoint of an MVP the point of cloud native is that you are leaning on the cloud provider to do what otherwise would be a big hassle.
That said if you are using a conventional architecture you can usually run it on a VM in the cloud.