So I figured it was about time I documented how to build persistent memory for AI agents using the databases you already know. Not vector databases - MySQL and Neo4j.

This isn't theoretical. I use this architecture daily, handling AI agent memory across multiple projects. Here's the schema and query patterns that actually work.

The Architecture

AI agents need two types of memory:

• Structured memory - What happened, when, why (MySQL)
• Pattern memory - What connects to what (Neo4j)

Vector databases are for similarity search. They're not for tracking workflow state or decision history. For that, you need ACID transactions and proper relationships.

The MySQL Schema

Here's the actual schema for AI agent persistent memory:

-- Architecture decisions the AI made
CREATE TABLE architecture_decisions (
    id INT …

People at Intel started the pmem library project some time ago, it’s open to the broader community at GitHub and  other developers, including Linux kernel devs, are actively involved.

While the library does allow interaction with an SSD using a good-old-filesystem, we know that addressing SSD through SATA or SAS is very inefficient. That said, the type of storage architecture that SSD uses does require significant management for write levelling and verifying so that the device as a whole actually lasts, and your data is kept safe: in theory you could write to an NVRAM chip, and not know when it didn’t actually store your data properly.

But there are other technologies, such as Memristor (RRAM) and Phase Change Memory …