Trying to make cloud infrastructure as simple as possible for ML engs. What do you think?

Follow the full discussion on Reddit.
Hey guys! Over the past few months, I talked to several ML engineers (mostly startup founders) and realised that one thing all of them disliked was setting up & managing cloud infrastructure for their backend or ML model. Although there are services like Render, for some reason, they all opted to do it hardcore with AWS / Azure / GCP. Not sure why, but anyways, these services do have a lot of overhead which make them tedious or difficult, especially for first timers.