Barcelona · Est. 2026

Memory infrastructure
for the agent economy.

The inference economy runs on context. Human syntax is not the native language of machines. Geometrical is the memory substrate on which agents operate.

arXiv Personal Episodic Identity Substrate (PEIS) — coming soon · GitHub AIX Protocol v0.2.0 — open source, Apache 2.0 · Patent P202631047 filed OEPM — July 2026 · Research AIX Protocol Paper — arXiv submission pending · Press Coverage and interviews — coming soon · arXiv Personal Episodic Identity Substrate (PEIS) — coming soon · GitHub AIX Protocol v0.2.0 — open source, Apache 2.0 · Patent P202631047 filed OEPM — July 2026 · Research AIX Protocol Paper — arXiv submission pending · Press Coverage and interviews — coming soon ·

01 / Thesis

The relational database made the software era possible.
Geometrical builds its equivalent for the agent era.

Agents need to know who they are working for — their goals, decisions, criteria, roles, and history — not just the last message in a thread. Today, that context is improvised, duplicated, and forgotten on every new conversation.

Geometrical builds the primitive that was missing: a persistent, structured, privacy-native memory substrate. Local by architecture. Machine-readable by design.


Four surfaces.
One core engine.

All surfaces share the same engine — they differ only in subject and governance model.

S-01

Self

Personal Memory Substrate

The episodic memory of the individual. Captures intention, decisions, and criteria — not just facts. Operates across roles: founder, parent, partner, learner. Zero interaction required.

In development

S-02

Soma

Institutional Memory Substrate

The same 9-dimensional model applied to collective entities. Organizational knowledge that survives team changes, restructuring, and time. Enterprise deployment.

Piloting with partners

S-03

Nexus

Permission Protocol

Governs what memory dimensions are shared between a Self and a Soma, and between Self instances. Federated graph with user-controlled perimeter definitions. Open source.

Roadmap

S-04

Axis

Operational Memory Substrate

Memory substrate for autonomous agents. Agents running within the Nexus permission boundary can read the relevant dimensions of a user's identity graph. Humans control what context is made available.

Roadmap

02 / Core Engine

What runs
underneath.

Five subsystems, shared across all surfaces. The architecture is what makes Geometrical infrastructure rather than a product.

ENGINE · 01

Knowledge Compiler Dedup LSH + JSON Distiller

Converts raw episodic signals into structured, deduplicated memory units. Locality-sensitive hashing eliminates duplicates before storage. JSON distillation produces machine-native context.

Active

ENGINE · 02

PII Protect Layer Privacy by Architecture

Sensitive entities are transformed before any LLM interaction. Privacy is a mathematical property of the data, not a policy applied after the fact. Patent filed.

Patented

ENGINE · 03

Semantic Version Control Git for memory

Every memory write is a versioned, staged commit. Beliefs are not overwritten — they are evolved, with full audit trail. Prevents belief corruption in long-term identity graphs.

Active

ENGINE · 04

Permission Engine Nexus governance layer

Controls what memory dimensions are readable by which surface or agent. Permissions are explicit, user-defined, and revocable. The boundary between what an agent knows and what it may know.

In Design

ENGINE · 05

Predictive Context Engine Role-routed retrieval

Infers the active role from behavioral signals and routes memory retrieval accordingly. No explicit declaration needed. Founder context and parent context do not bleed into each other.

Active

03 / Self

Self

Personal Episodic Memory Substrate

"You don't remember how you breathe.
You won't remember how agents know what they know.
That is Self."

Self is not an app. Not an assistant. Not a hub. It is the memory layer that makes every agent you interact with coherent over time — across roles, across devices, across contexts.

It captures what matters: intention, criteria, decisions, patterns. Not transcripts. It operates passively across eleven ingestion sources. It requires zero ongoing interaction.

Architecture Local-first by design. Nothing leaves the device without explicit, user-defined permission. Not a policy — a mathematical constraint.
Roles Infers multiple simultaneous roles — founder, parent, learner, partner — from behavioral signals. No declaration required.
Format .aix — a 9-dimensional identity snapshot. Machine-readable, portable, spatiotemporally traceable. The PDF of the agent era.
Hardware Runs on consumer devices today. iPhone 15 Pro, iPad Pro M1, Mac M4. Local LLM inference — no cloud dependency in the core loop.
Recovery Self doesn't live on your device. It lives in your personal ecosystem. Encrypted .aix backup + peer-to-peer sync across your devices. Recovery phrase for key reconstruction.
Privacy PII is transformed before any model interaction. No fine-tuning. No equivalence tables transmitted. Privacy is in the mathematics, not the terms of service.

Research & press

04 / Publications

05 / Contact

Build something that lasts.

We are building with a small number of partners. If you are working on the agent stack, investing in infrastructure, or want to pilot Soma inside your organization — reach out.

hello@thegeometrical.com