Rethinking Databases for Humans and AI Agents
Marco Bambini
SubscribeSign in
Rethinking Databases for Humans and AI Agents<br>Databases Were Built for Engines. It’s Time to Build Them for Humans.
Marco Bambini<br>Jul 01, 2026
Share
Databases are one of those technologies that seem to survive every technological revolution while becoming even more important each time.<br>First, there was SQL. Declared dead countless times, yet still at the center of much of modern software. SQL made it possible to organize enormous amounts of information efficiently and reliably, but it always imposed a precise model: tables, columns, relationships, indexes, and queries that developers needed to understand in detail.<br>Then came the NoSQL wave, and databases like MongoDB changed the perspective. No more rigid schemas, but flexible JSON documents that better matched how developers represented data in modern applications.<br>And yet, one thing never really changed.<br>To store information in a database, you still need well-defined operations such as INSERT or UPDATE. To retrieve information, you still need to know columns, keys, structures, or implementation details of the underlying engine.<br>In other words, databases still speak the language of the engine, not the language of humans or AI agents.
Let’s think about a very simple example.<br>Imagine we want to store the following information:<br>“Marco met Andrew in San Francisco to discuss a possible integration between SQLite AI and A51 Robotics.”
In the SQL world, we would probably need to create something like this:<br>CREATE TABLE meetings (<br>id INTEGER PRIMARY KEY,<br>person1 TEXT,<br>person2 TEXT,<br>city TEXT,<br>topic TEXT<br>);<br>INSERT INTO meetings<br>VALUES (<br>1,<br>‘Marco’,<br>‘Andrew’,<br>‘San Francisco’,<br>‘SQLite AI integration with A51 Robotics’<br>);
And later, to retrieve that information, we would need to know the database structure and write queries such as:<br>SELECT *<br>FROM meetings<br>WHERE person2 = ‘Andrew’;
Or:<br>SELECT *<br>FROM meetings<br>WHERE topic LIKE ‘%A51%’;
SQLite-Memory starts from a completely different idea.<br>The information can simply be stored by writing a Markdown file:<br># Meeting Notes<br>Marco met Andrew in San Francisco<br>to discuss a possible integration<br>between SQLite AI and A51 Robotics.
No tables.<br>No columns.<br>No schema to design.<br>And most importantly, retrieval becomes natural.<br>You can simply ask:<br>“What did we discuss with Andrew?”
Or:<br>“What discussions did we have regarding A51 Robotics?”
Or even:<br>“Do you remember the meeting in San Francisco?”
Without knowing the name of a column, without knowing how the information was stored internally, and without writing SQL queries.<br>Because humans do not think in tables.<br>And AI agents do not either.<br>We believe the future of databases should be closer to how people actually think: natural language, context, and semantic memory.<br>SQLite-Memory was created precisely around this idea.<br>Instead of forcing users to design schemas and queries, SQLite-Memory lets them build persistent memory from simple Markdown files. The information is processed by a specialized parser, automatically structured, and made semantically searchable.<br>Search is no longer based solely on columns or keywords but on the meaning of the information itself.<br>Because we believe the future of databases is not just about storing data, but about making information understandable and accessible in the most natural way possible.<br>SQLite-Memory is available today as a native SQLite extension and is also built directly into the SQLite Cloud platform. We are also working on a new PostgreSQL version to bring the same paradigm to existing PostgreSQL infrastructures and AI-native applications.
Memory becomes truly powerful when it can be shared.<br>A personal memory is useful. A collective memory is transformative.<br>This is where SQLite-Sync comes into play.<br>SQLite-Sync allows semantic memories to be synchronized across devices, users, and AI agents while remaining fully offline-first. Knowledge created on a laptop can instantly become available on a phone, shared with a team, or incorporated into the working memory of multiple AI agents collaborating on the same tasks.<br>Traditional synchronization systems were designed around rows and records. Semantic memory introduces a different challenge: synchronizing knowledge, documents, and evolving context without losing meaning or creating conflicts.<br>To solve this problem, SQLite-Sync uses a new CRDT algorithm specifically designed for Markdown-based semantic documents. Instead of treating content as opaque text blobs, it synchronizes knowledge at the block level, making collaborative editing, agent-to-agent memory sharing, and distributed semantic knowledge bases possible even in unreliable or completely offline environments.<br>We believe this combination of semantic memory and offline-first synchronization represents a foundational building block for the next generation of AI applications, where humans and agents continuously create, refine, and share knowledge...