Independent Research & Technology Lab

Autonomous Intelligence.
Cognitive Architectures.
Resilient Systems.

Duara Cortex is a United Kingdom research organisation studying the architectural foundations of agentic computing: how autonomous systems represent goals, coordinate with one another, maintain state over long horizons, and recover from their own failures. Our work sits at the intersection of cognitive science, distributed systems, and software engineering, and proceeds from a premise we take seriously — that these questions require sustained, unhurried inquiry, and are poorly served by research programmes structured around commercial delivery cycles. Duara Cortex was incorporated as a non-profit specifically to remove that constraint.

Explore Research Areas Read the Charter

// SPECIFICATION ACTIVE LAB

Organisation

Duara Cortex Limited

Entity Form

Company Limited by Guarantee (Not-for-Profit)

Jurisdiction

England and Wales

Inquiry Focus

Agentic Architectures & Distributed Systems

Asset Lock

Constitutional / Irrevocable

Official Code Registry

github.com/Duara-Cortex ↗

[02 // FOUNDATIONS]

Mission, Governance & Advisory Work

Duara Cortex was established to separate a specific kind of work — long-horizon research into agentic and cognitive systems — from the commercial pressures that ordinarily shape it. As a Company Limited by Guarantee, the organisation has no shareholders and no equity to protect; it exists to hold intellectual property, enter into contracts, and conduct research in its own name, with any operating surplus directed back into the work rather than distributed as return.

That governance structure is deliberate rather than incidental. Much of what makes agentic and cognitive-architecture research difficult — reflective planning, long-term memory, multi-agent consensus — resists the kind of short-cycle, shippable-feature research that a commercially incentivised lab tends to favour. Removing the shareholder relationship removes that pressure at the source.

Foundational Principles & Advisory Model

01 PRINCIPLE

Intellectual Autonomy

Operating as a Company Limited by Guarantee, Duara Cortex reinvests the entirety of its operational surplus into computing infrastructure, research tooling, and the pursuit of open questions, rather than into returns for owners who do not exist.

02 PRINCIPLE

Open Inquiry & Rigour

Findings are held to a standard of empirical reproducibility and formal specification, and are published — including negative or inconclusive results — so that the underlying claims can be examined and, where warranted, contested.

03 PRINCIPLE

Applied Software Craft

Theoretical mechanisms are not considered complete until they have been implemented as production-grade software: tools an engineer outside the organisation could adopt, depend on, and reason about.

04 PRINCIPLE

Applied Research Consultancy

Duara Cortex undertakes a limited number of advisory and technical consulting engagements with organisations designing or operating agentic systems, cognitive architectures, or knowledge-graph infrastructure. These engagements are structured deliberately: they are a means of testing our research against real operating conditions, and any income they generate is subject to the same asset lock as the rest of the organisation — reinvested in research, not extracted as profit. Consultancy is a channel through which our work meets practice, not a separate commercial arm.

[03 // INQUIRIES]

Core Research Areas

Our research is organised around three interdependent problems. We treat them as interdependent deliberately: a cognitive architecture that cannot be run reliably is not useful, and a reliable runtime with no capacity for reasoning is not interesting. The three areas below are where those two ends meet.

INQ-01 RESEARCH PROGRAMME

Cognitive Architectures & Agentic Workflows

Multi-Agent Consensus Memory Synthesis Reflective Planning Distributed Coordination

This is the study of how systems composed of multiple autonomous agents should coordinate: the protocols by which they reach consensus, allocate work, and resolve disagreement without constant human arbitration. It extends into long-horizon memory synthesis — how an agent should decide what to retain, what to discard, and what to revisit — and into reflective planning, where a system evaluates and revises its own intermediate reasoning rather than committing to a single, unexamined plan.

INQ-02 RESEARCH PROGRAMME

Deterministic Systems & Runtime Engineering

Go Python TypeScript Concurrent State Sandboxing Low-Latency Runtime

Reasoning is only as trustworthy as the infrastructure executing it. This area covers the construction of resilient backend architectures, principally in Go, Python, and TypeScript, with particular attention to state management under concurrent agent activity, execution safety through sandboxing, and the low-latency infrastructure that long-running agentic processes require to remain economically viable.

INQ-03 RESEARCH PROGRAMME

Knowledge Graphs & Semantic Synthesis

Relational Models Formal Verification High-Dimensional Search Retrieval-Augmented Reasoning

Unstructured information is not, by itself, reasoning material. This area addresses how such information is structured into relational models that can be formally verified, queried, and traced back to source — including work on high-dimensional search, retrieval-augmented reasoning, and the context-window and relevance-ranking problems that arise once a knowledge graph is put to use inside an agentic workflow.

[04 // REPOSITORIES]

Applied Software & Projects

Duara Cortex treats its published software as the primary evidence of its research, not as a secondary output. All formal releases are maintained under the organisation's GitHub account: github.com/Duara-Cortex

The organisation is at an early stage: formal releases under the Duara Cortex account are forthcoming, as existing exploratory work is formalised, documented, and migrated into the organisation's public repositories. That existing body of work already gives a reasonably direct sense of the research programme, and includes:

REPOSITORY MIGRATION

Exploratory codebases are actively being consolidated and prepared for public Apache-2.0 release under @Duara-Cortex.

Exploratory Implementation

Self-Hosted Knowledge Vault

A system for aggregating and structuring personal and organisational knowledge — a practical testbed for the semantic-synthesis research described above.

Markdown Knowledge Graph Local-First

Exploratory Implementation

Orchestration Daemon

High-precision daemon for scheduling and executing agentic routines within precise time windows — early infrastructure for the coordination problems in agentic workflows.

Go Concurrency Agent Scheduling

Algorithmic Prototype

Deterministic Hashing Construction

Algorithmic construction built on recursive geometric expansion — an example of deterministic-systems research applied to a concrete cryptographic problem.

Cryptography Deterministic Algorithms Geometry

Research Testbed

Long-Term Memory Layer

A graph representation mapping notes and their links weighted by usage and connection density — a direct precursor to our published work on memory synthesis.

Graph Theory Memory Synthesis Vector Semantics

Active Tooling

Remote Orchestration API

An API for uploading, managing, and triggering agentic skills and tools from external harnesses.

TypeScript APIs Skill Harness

PUBLIC REGISTRY

Duara-Cortex on GitHub

Follow our public repositories, issues, open RFCs, and benchmark pipelines as implementations are formalised.

Explore repositories →

[05 // CONSTITUTION]

Governance & Operational Model

Duara Cortex Limited is incorporated in England and Wales as a Private Company Limited by Guarantee. This is a legal form typically used by organisations without a profit-distribution motive: rather than share capital and shareholders, the company is governed by members whose liability is limited to a nominal guarantee, and it carries no mechanism for returning surplus to private individuals.

Central to this structure is an asset lock: a binding constraint, written into the company's governing documents, that its assets and any surplus it generates may be applied only in furtherance of its stated purpose — research and public technological advancement — and may not be distributed, extracted, or diverted for private commercial gain, including in the event the company is wound up. This is not a marketing claim but a legal fact about the organisation's constitution, and is the mechanism that allows Duara Cortex to describe itself, without qualification, as not-for-profit.

Binding Asset Lock

Statutory restriction preventing the distribution of dividends, capital gains, or wind-up balances to private interests.

No Share Capital

Eliminates equity ownership pressures, ensuring research timelines are driven by empirical discovery rather than quarterly milestones.

Full Surplus Reinvestment

100% of consultancy revenues and grants are channeled into computing power, research tooling, and open software infrastructure.

[06 // DISPATCH]

Inquiries & Collaboration

Duara Cortex welcomes contact from academic researchers working in adjacent areas, from open-source contributors interested in the software described above, and from organisations considering an advisory or consulting engagement. Correspondence is read directly by the research team rather than routed through a general inbox.

Academic & Technical

Joint empirical investigations, paper reproductions, or open architectural critique.

Advisory & Systems Review

Targeted consulting engagements for agentic runtime safety and semantic knowledge architectures.

Direct Channels RECEPTIVE

Electronic Mail

contact@duaracortex.com

Correspondence & Registered Office

Duara Cortex Limited
66 Paul Street
London, England, EC2A 4NA
United Kingdom

Codebase & Research Issues