Inside the ideas

How the pieces
work together.

Explore the responsibilities, information flows and design choices behind the work. Each guide connects an idea to a concrete example, its current scope and the questions still to resolve.

Follow the architecture behind the work.

Explore the responsibilities, information flows and design choices in each system, then open the retained source views where more detail is useful.

Optional: retained sources and alternative architecture views

Retained source and comparison views

Original source reviews and alternative context, container and component views remain available for deeper reading. Each retains its own date, scope and interpretation.

Conduir / Role Fabric

Work custody, role execution, scheduling, memory adapters and governance boundaries.

Full architecture and diagrams · C3 alternative

Cortex

All memory representations, retrieval, provenance, memory care and current source limits.

Full architecture and diagrams · C3 alternative

GraFOSa

Graph-based software understanding, composition, mathematical analysis and optional learning paths.

Full architecture and diagrams · C3 alternative

Original architecture records · All 18 current / C3 comparisons · Explore the complete research suite · Explore DocuMind

Systems & engineering

Architectural programme

Conduir: architecture for work that compounds

Conduir is a programme of cooperating capabilities. Context, coordination, action, review and governance have distinct responsibilities. The architecture asks how they can form a continuing journey, so useful work and the reasons behind it do not disappear at the next handover.

Explore the architecture ↗

Memory architecture and implementation work

Cortex: memory that keeps context and purpose together

Cortex separates capturing information, preserving its context, retrieving relevant records and caring for the resulting memory. Memory retrieval provides evidence to a task; it is a different mechanism from changing the parameters of a neural model.

Explore the architecture ↗

Structural analysis tooling

GraFOSa: understanding the relationships around a change

GraFOSa represents code structure as related entities rather than isolated files. Questions about a change can then follow those relationships. An isolated graph alternative offers a place to examine a proposal while keeping structural analysis distinct from executing and testing software.

Explore the architecture ↗

Idea-lineage research

Manthan: ideas with a visible history

Manthan separates generating contributions, relating them, preserving their origins and deciding which deserve further attention. Its lineage record is useful even when the quality of an idea remains unsettled. Selecting consistently useful ideas is a separate question from producing many of them.

Explore the architecture ↗

Review mechanism

AI Review Panel: making different perspectives useful

The panel separates review from adjudication. Different models examine the same subject; their findings are then challenged, consolidated and checked against that subject. The useful output is a traceable set of dispositions, not a vote that declares the work correct.

Explore the architecture ↗

Engineering practice

AI Engineering Workforce: how the work is organised

This is an operating approach around engineering, not another platform successor. It separates work planning, implementation, independent review and completion decisions. The product architecture describes what is being built; the workforce approach describes how contributors work on it.

Explore the architecture ↗

Shared-environment research

Conduir Spatial: working within a shared scene

Spatial work combines a shared scene, participant interaction, proposed actions and a record of change. A shared view is only part of collaboration: participants also need to know who proposed a change, whether it was accepted and what context should survive.

Explore the architecture ↗

Earlier platform work

Maestro: coordinating specialist intelligence

Maestro approached coordination through specialist roles, workflow structure and continuing session context. The later Maestro Platform broadened the engineering foundation; Conduir carried forward implementation and architectural work from that history.

Explore the architecture ↗

Learning & creation

Research implementation

Modular ANN: learning in parts

Investigate whether intelligence can acquire useful capabilities through smaller, explicitly controlled changes, while preserving the ability to examine what changed.

Explore the architecture ↗

Research implementation