4 databases for vibe coders

4 databases for vibe coders. I think you have heard of Supabase, but that's one type of database. Choosing the right type of data format is what's going to prevent your app from crashing when you get a lot of data or users. And your software engineer friends will tell you nobody wants to migrate a database halfway through. So this video, let's learn about four types of databases. First, there's Postgres and an example is Supabase. This is your standard Excel sheet looking database with rows and columns. What you do is you set up an index on whatever columns you're doing lookups on, and you can generally slice and dice just like an Excel sheet. For the same type of data, you can also use NoSQL databases and an example is Firestore. This type is more scalable than Postgres, but you can't do the general slice and dice. You have to specifically know how you're going to do lookups and you have to index on that column. If you can manage this, you can be cheaper and faster. If you're doing things like RAG, you need a vector database like Pinecone. One way to think about vectors is imagine you have millions of documents and each document represents a point in space. Whenever somebody has a question about those documents, that question also becomes a point in space, and then we just choose whatever documents are closest to that question point. That's a vector database which serves a specific purpose. The last type of database I can talk about is called Elasticsearch. This is a combination of vector which we just talked about and also general word search. And this type of database will power search engines. Use the right type of database for your product.