MY MEMORY

Your memory should belong to you.

User-owned continuity infrastructure for long-running AI-assisted work.

The problem

A serious project can outlive a chat, a session, a model and eventually an AI provider.

When that happens, decisions become separated from their reasons, sources disappear from conclusions, commitments drift out of view, stale information competes with current state, and the user is asked to reconstruct the same project history again.

A successor AI should not have to guess what happened before it arrived.

What MY MEMORY does

MY MEMORY is an external project-memory layer focused first on Project & Decision Memory. It preserves the parts of long-running work that must survive beyond a single conversation or provider.

Project → Event → Decision → Reason → Person → Document → Commitment → Current State → Source

The AI participates in memory capture. The user owns the storage.

The system is designed to distinguish retained history from accepted current truth. Facts, decisions, reasons, assumptions, uncertainty and third-party statements should not collapse into one undifferentiated archive. Provenance matters, and later correction should supersede earlier state without pretending that the earlier record never existed.

Technical depth

MY MEMORY is not a generic LLM wrapper, a web portal or a service business. Its technical core is a provenance-aware continuity architecture for maintaining explicit project state across sessions, models and AI providers.

The implemented mechanics distinguish current, stale, superseded, conflicting and uncertain information; preserve decisions together with reasons and sources; and use explicit human-reviewed state transitions rather than treating opaque conversation retrieval as authoritative memory. Recovery work has already exercised clean, adversarial and cross-provider conditions, including rejection of stale state and explicit acceptance of a fresh return.

That technical depth is established by implemented continuity and recovery mechanics and by the bounded tests described below. What is not yet established is a legally defensible moat: patentability, freedom to operate and the long-term difficulty of reproducing the architecture remain open questions.

Evidence so far

What this evidence does not establish

• Product-market fit or repeated target-user adoption.
• Production security, scale or multi-user readiness.
• Provider-neutral writable interoperability across all AI systems.
• Long-duration reliability under production conditions.
• Patentability, freedom to operate or a proven defensible IP moat.

The current evidence supports bounded continuity and recovery behavior. It is not presented as proof of a finished product.

Current product — now

Project & Decision Memory for people running real, multi-stage AI-assisted projects.

The initial target is work where losing context has a real reconstruction cost: founders, researchers, writers, creators, consultants and other people whose projects continue across many sessions and tools.

Long-term direction

The larger idea is personal continuity: memory that remains with the person as years pass, projects change and AI systems come and go.

The long-term aim is not merely to make one AI conversation remember another. Continuity between sessions, models and providers is an enabling mechanism. The deeper value is to help a person preserve and later recover the structure of their own history: what happened, what they knew at the time, what they decided, why they decided it, what sources shaped that decision, what changed later and which version of the record is current.

In that direction, AI becomes an interface to user-owned memory rather than the owner of it: an agent that can help capture, connect, retrieve and explain records while the durable memory remains external to any single AI provider. The same continuity principles may later extend into organizational, research, creative and other long-lived knowledge environments where provenance and current state matter.

Future research. A separate research horizon is personal cognitive continuity over long periods of life, including ordinary human forgetting and, separately, whether such user-owned memory structures could one day be useful in research involving early memory decline or early-stage dementia. That medical or cognitive-health branch is not the current product. MY MEMORY is not a medical device, diagnostic system, dementia treatment or clinical intervention, and no clinical benefit is claimed; any such application would require dedicated research, safety work and appropriate validation.

Current stage

• Working prototype.
• Pre-revenue.
• Pre-seed.
• Operational recovery evidence established in bounded tests.
• Real target-user value and repeated adoption still under validation.

Next milestone

Turn the validated recovery protocol into a simple daily-use MVP, automate meaningful source capture, improve relevance-aware retrieval, and test whether real target users repeatedly choose the product because it materially reduces reconstruction work, stale-history errors and continuity loss.

Funding

MY MEMORY is currently preparing a pre-seed round in the approximate range of €80,000–€100,000 to fund roughly twelve months of focused founder, product and user-validation work. The purpose of the capital is to create sustained development time and move from validated recovery protocols toward a usable product with real target-user evidence.

Target outcome: a daily-use MVP, measurable repeated use by external target users, and a stronger evidence base for the value of continuity.

Founder

Nikolai Kostyshev / Ashraellen — independent founder, based in Poland.

MY MEMORY grew from the practical need to preserve continuity across the founder's own long-running AI-assisted projects, where decisions, reasons, sources and current state had to survive beyond individual sessions and tools.

Contact

For investment, product validation or relevant collaboration:

ashraellen.live@gmail.com

— mark of presence