1
The technology landscape is changing rapidly across the technology adoption (devops, security, etc.), people management (all-remote-is-a-reality!); how do you remain on top of your game?
Are there any considerations on limiting the public exposure of internal APIs? Or have you seen any successful implementation of a multi-tenant API where the product and third-party apps both use the same underlying API infrastructure?
In most of the cases, the predictions are meant to be delivered via a REST API. One of the approaches is to pre-compute predictions -> store in DB -> serve via API.
Do you have a better pipeline for your production models?