People are mostly running a very local stack for memory I think creating really solid custom cloud based memories or brains for LLMs is about to explode in demand well it currently already is kinda ai startup aitools secondbrain techtok

Do you wanna know the weirdest thing? Companies are ripping out their MCP servers and then replacing it with a $20 database. Go and look this up. Everybody wants to connect their data and software and shit to Cloud. The idea of creating a second brain for your company sounds really cool. Well, go and set up a system that looks like this. They will whip up a couple Cloud connectors, shoot Cloud a prompt like this. Then it will call like 14 different tools to figure out the answer. Then it will usually fuck up a little bit. Then most companies will give up there trying to create a second brain. Wait, there are certain companies that have figured this out. This is the creative approach I'm seeing people take. This is the prompt that they're using to do it. They will sign up for a database like this, set up some workflows that'll just sync all their data to this database. The tables in this database would probably look something like this. Then they'll set up a bunch of tools, kind of like this one. What used to take like seven tool calls is now just one single tool call. It'll get all of the activity from all of these things at once. If you're gonna create a second brain for your company or somebody else's, do something like this. That's why I gave you this prompt, bro. Just give this a rip-a-roni. Yeah, people are still kind of figuring out what and what doesn't work. That's what I'm spending all my time on at least. Oh, you will be kept updated.