- Technical deep dives
Devlog: How Eigen Mesh is actually built
An in-depth architecture tour of the Eigen Mesh memory engine — the three-tier pipeline from capture to recall, hybrid retrieval with weighted fusion and an LLM rerank, nightly consolidation, one Postgres doing three jobs, and the rules we don't break.
- Why personal AI memory
Your AI remembers everything you tell it. Then it sells it.
Why ChatGPT and Google AI's built-in memory is a trap, and what a memory you actually own looks like.
- How Eigen Mesh works
Eigen Mesh in 60 seconds
A quick, plain-language explainer of what Eigen Mesh does and why it's different from the usual note apps.
- Anti-Big-AI
Why I'm building Eigen Mesh
A founder's note on where Eigen Mesh came from, and why your thoughts and your agents' thoughts sit side by side in it.
- Technical deep dives
Self-hosting Eigen Mesh in 5 minutes
A walkthrough of running Eigen Mesh on your own server with Docker Compose — what the stack looks like, what you bring, and where to look next.
- Use cases
"Remember that restaurant I mentioned?" — AI that actually does
A small, everyday use case for AI memory — finding the place you told your assistant about three weeks ago, without scrolling.
- Anti-Big-AI
The AI memory war: who's winning and why you should care
A field guide to the AI memory land grab — the players, the tactics, and the one bet they're all making about you.
- Why personal AI memory
"Hey ChatGPT, remember this" — but who's really listening?
The memory features in consumer AI tools turn your private thoughts into product data. Here's what's actually happening.
- How Eigen Mesh works
Your AI memory, explained like you're five
A plain-language walkthrough of how Eigen Mesh's memory actually works, no jargon required.
- Technical deep dives
How Eigen Mesh handles embeddings without vendor lock-in
Why Eigen Mesh stores embeddings in Postgres with pgvector, keeps them DB-only, and lets you bring your own embedding model via any OpenAI-compatible gateway.
- Use cases
Remembering your kids' allergies, schedules, and stories
A use case for parents — the details you don't want to forget, kept in your own words.
- Anti-Big-AI
Google, OpenAI, and the race to own your context
Two companies are spending billions to become the place your AI memory lives. Here's what each one wants, and why you lose either way.
- Why personal AI memory
Switch AIs without starting over
Why Eigen Mesh keeps your memory in a layer of its own, so you can move between models and tools without losing anything.
- How Eigen Mesh works
Less than a coffee a month
Why a simple €3.99/month subscription beats opaque usage-based pricing for an AI memory you actually use.
- Technical deep dives
MCP + Eigen Mesh: giving every AI tool a shared memory
How Eigen Mesh exposes memory over the Model Context Protocol so Cursor, Claude, and other MCP clients share one second brain instead of siloed context.
- Use cases
Your therapist remembers. Your AI should too.
Why continuity of personal context across sessions matters — and why it shouldn't require handing your inner life to a chatbot company.
- Anti-Big-AI
Why "free" AI isn't free
Free-tier AI products are not gifts. They are a pricing strategy, and the price is almost always your context. Here's the honest math.
- Why personal AI memory
The AI that forgets: why your context shouldn't belong to a company
Vendor lock-in isn't an accident in consumer AI memory — it's the product. Here's why that's a problem.
- How Eigen Mesh works
Open source doesn't mean complicated
The "open source = hard" myth, and why Eigen Mesh ships a managed version alongside the self-hostable one.
- Technical deep dives
Contributing to Eigen Mesh: where we need help
An honest look at the open areas where Eigen Mesh welcomes contributors — docs, evals, MCP clients, i18n, and self-host UX — and how to start without burning out.
- Use cases
Stop re-explaining yourself to every AI tool
The case for a memory that follows you between Cursor, Claude, and the next thing — instead of one that dies with each app.
- Anti-Big-AI
Open source AI infrastructure: the quiet revolution
The flashy part of AI is closed. The part that actually runs the world is quietly going open source — and memory is the next layer to fall.
- Why personal AI memory
What happens to your AI conversations when the company goes down?
When your AI memory lives with a vendor, their outage is your amnesia. Open source is the insurance policy.
- How Eigen Mesh works
How your memories stay private (even from us)
The concrete technical and operational choices that keep your Eigen Mesh memory yours — encryption, isolation, no training, EU hosting.
- Technical deep dives
The storage layer: Postgres + pgvector and why it works
Why Eigen Mesh uses one PostgreSQL instance with pgvector and Apache AGE for relational data, vector embeddings, and the knowledge graph — and what that buys you.
- Use cases
The solo founder's memory bank
Decisions and context kept in one place you own, available to every AI tool you use to run your company.