I like the feature where we can build Gen AI application directly on EDB Postgres AI for CloudNativePG. The scaling of AI models within the database is another aspect I appreciate. The Unified workloads feature, allowing for both traditional transactional queries and AI tasks on the same platform, stands out to me. Additionally, I like the hybrid data manager feature. With these capabilities, it's useful for handling growing chatbot data and supports vector search, which is widely used for AI applications. Review collected by and hosted on G2.com.

It's very difficult to create initial setup and configuration database for AI workloads due to improper documentation, so I'd like to improve the document's content. There are some advanced features in EDB, but due to lack of documentation, we are not aware of how to use them. The PostgreSQL uses one primary writer node with multiple read replicas which becomes a bottleneck for very high write workloads. It should follow similar to Google Spanner so that we scale horizontally. It's very hard to set up initially without any expert in PostgreSQL. The challenge we faced is how to configure replications and containers, and it's very difficult to configure security as well. Review collected by and hosted on G2.com.