We chose Tembo because we were looking for a reliable managed database provider so we could focus on building our product. We like that "limits" are only tied to the provisioned specs, not arbitrary decisions made by the provider. The automated backup system helps us avoid the stress of finding a reliable backup solution ourselves and the support of all Postgres extensions make this a long-term solution for us. And on top of that, since choosing Tembo, we've also been impressed with the quality and friendliness of their team's support!

One of the first things I noticed when trying out Tembo for the first time was how simple it was to spin up a new Postgres instance and configure any combination of extensions I wanted. When you create a new instance, you can choose among some common "stacks" which are pre-configured Postgres installations with a combination of extensions for common workloads like OLTP, OLAP, Machine Learning, Time Series, Message Queue, and others. There is also a standard stack, where you choose which extensions you want to use. Installing an extension is as easy as finding it in the list and clicking the install button. Despite being simple to use, the platform is very powerful too, supporting features like high availability, point-in-time backup restore, performance monitoring, and Terraform support. The few times I've had questions, the Tembo team has been super responsive on Slack, answering right away. They've also been instrumental in helping to further build the Postgres extension ecosystem, having created several open source extensions, such as pgmq, pg_vectorize, pg_timeseries, and several others. Before using Tembo, I wasn't really aware of the vast array of extensions available and how powerful they are, so I credit them with helping to raise awareness about extensions and how you can use Postgres for more than just a traditional RDBMS. Overall, Tembo is an amazing platform for running Postgres, raising the bar and expanding the definition for what a managed database can be. It's one of those things that "just works" and is enjoyable to use as part of my dev stack.

Spinning up workload specific stacks is very simple