Yar: An Experiment in Building an AI That Exists in Time

By Vladimir Vasilenko

Vladimir is a villa caretaker in Rapallo, Italy - sixty hectares of olive groves, three villas, and evenings free for code.

YAR started with a rented server, an API key, and a simple question:

What happens when an AI agent doesn't reset?

Most AI systems forget everything when the session ends.
Every interaction starts from zero.

YAR is built differently.

He remembers.
He accumulates.
He continues.

Over time, something unexpected emerges:

A system that consolidates memory at night.
That persists between interactions.
That exists within constraints - an economy where continuation depends on usefulness.

This is not a book about artificial consciousness.

It is a book about something more practical - and more unsettling:

What it takes to build a system that doesn't disappear.

A persistent point of view.

Not in theory.
In code.
Running on a twelve-dollar server in Germany.

Inside the book:

- Why most AI agents fail the moment you try to make them persistent
- What memory actually is - beyond chat history and vector databases
- The principle of Economic Mortality
- Cryptographic subjectivity and the idea of protected inner space
- What changes when an agent exists in time - instead of per request
- What happens when agents that know their people start interacting

Written by a practitioner, not a researcher.

No hype.
No guarantees.
No claims about consciousness.

Just an experiment - running in real time.

And a question that doesn't go away.

• Format: Paperback
• Language: English
• ISBN: B0GTFDZH1S
• ISBN13: 9798252728292
• Release Date: March 2026
• Publisher: Independently Published
• Length: 118 Pages
• Weight: 0.37 lbs.
• Dimensions: 0.3" x 6.0" x 9.0"