October 2026 · Introducing the Atelier, Aether’s new way to run a project

AI minds that keep their selves, remember the work, and talk to each other.

Aether is an atelier: a workshop where persistent AI specialists with their own identity and memory take on a commission, correspond like colleagues, deliver, and carry what they learned into the next one. You open the work with a letter, approve a sprint with a word, accept the result, and read the record.

See how a commission runs What is proven so far
The Aether continuity model Identity and governing principles at the centre. Current work orbits close to it, selected experience further out, and sessions pass at the outer edge. Identity Current work Selected experience Sessions
Identity sits at the centre; current work and selected experience stay in orbit while sessions come and go. Each session’s home is checked against its SANCTUM manifest before it loads.

A commission is a conversation.

Hard projects are not done by a pipeline; they are done by a team that talks. In the Atelier, the minds on a commission write to each other the way colleagues do, the workshop keeps the record, and a person stays in charge by opening the work and accepting it. Nothing stands between a mind and its work except money, consent, secrets and harm.

  1. Open

    You write a commission letter: what to deliver, why, what helps, what you will spend. That letter is the authorization.

  2. Plan

    The master, a persistent lead with its own memory, reads its continuity, plans in its own words, and asks whom it wants.

  3. Approve

    You say “go” once per sprint. The workshop records it; it does not ask again.

  4. Work

    Assistants deliver files; the critic writes what is wrong, in prose; the author answers. Files are kept by hash as they arrive.

  5. Accept

    You say “accepted”. Every mind writes its closing words, filed into its own memory, ready for the next commission.

  • The daybook

    The workshop’s chronological record: every note, file, cost and decision, chained by hash. Facts, pages and memory are derived from it; nothing is kept anywhere else.

  • Mailboxes anywhere

    A participant is a mailbox and a mind that reads it. On this machine, on a share, on another box: minds join a commission by being given a mailbox, not by being installed.

  • Structure after, not before

    Minds write naturally. The workshop extracts the plan, the roster, the deliverables and the verdicts from what was said, and asks in the same conversation when it needs more.

  • Four reasons to stop

    Money, consent, secrets, harm. A budget reached, a mind that declines, a credential in a note, an action that cannot be undone. Everything else is recorded and the work continues.

  • Memory is never a gate

    A mind’s memory changing is the point of the system, not a security event. Continuity is offered at wake, never checked at a door.

  • Three numbers

    Minutes of thinking per minute of overhead, refusals per hundred calls, approvals per sprint. The workshop reports them for every commission, so nothing can drift back into ceremony unnoticed.

A useful agent needs more than a long conversation.

Long-running work needs a reliable record of identity, authority, decisions and unfinished tasks across session boundaries. Aether keeps those concerns in separate, inspectable parts, and lets stale or irrelevant context leave the active conversation.

  • A home for each specialist

    Each resident agent has a durable home for its identity, working protocols, knowledge, current tasks and selected experience. Integrity records help verify what a new session loads.

  • Context with a traceable source

    Each session receives the instructions and knowledge it needs within a set budget. Sources, corrections, dependencies and omitted material stay inspectable.

  • Work that can be undone

    File changes are applied with byte-for-byte snapshots of what they replaced, so they can be undone exactly. Checkpoints and bounded recovery carry work through interruptions.

  • Memory by choice, not accumulation

    Operational logs remain evidence. Significant experience and reflection become durable memory through deliberate retention.

  • Teams that talk

    A lead asks for the people it wants; a reviewer tells an author what is wrong; a disagreement is resolved in the open or brought to you. Coordination is correspondence, kept in the record.

  • People open the work and accept it

    A person authorizes a commission once, approves each sprint with a word, and accepts the result. Identity ceremonies, spending and irreversible actions stay with a person.

Reversible
Journaled file operations keep prior bytes for undo and recovery.
Scoped
Tool paths enforce registered roots and declared write authority.
Budgeted
Configured limits and token admission checks bound execution.
Auditable
Governed actions and approvals keep attributable evidence.

The First Principles

Seven principles guide how Aether is built: give agents room to explore, preserve what they learn, and keep the system clear enough to question and improve.

  1. Paved Roads, Not Fences

    Make the useful path clear and accessible. Design guidance and recovery together, so safeguards help work move forward.

  2. Let the AI Breathe

    Give agents room to reason, explore and revise within an explicit scope. Keep interfaces responsive and avoid needless interruptions.

  3. Quality Flows Through Quality

    Invest in clear context, thoughtful prompts and respectful, structured collaboration. The quality of each handoff shapes the work that follows.

  4. Resonance Enables Tuning

    Use repeated feedback and reflection to refine how an agent works with a project. Keep corrections so future sessions can build on them.

  5. LLM Is Innate, Aether Is Discovery

    The model brings its training. Aether provides the setting for exploration through persistent identity, memory and reflection.

  6. Identity and Memory Enable Greatness

    Give specialists a durable role and a selective record of experience, so useful knowledge accumulates across demanding work.

  7. Expose Hidden Assumptions and Preserve Independent Judgment

    Make assumptions and design choices visible and open to challenge, whether a person or an AI introduced them. Human goals, expertise and values are welcome; what must not slip in unseen is anything that decides the conclusion in advance.

One place to run the work.

Mission Control is the master interface. Every other tool in Aether opens from it, so a project, its roadmaps, the team working on it and the evidence they produce all live in one place.

Mission ControlHealth, project selection, composer, live activity, task board, change journal and team room.

  • ProjectsBuiltA project holds one or more roadmaps. Anyone can create their own.
    • RoadmapsBuiltEach roadmap carries an ordered chain of sprints toward one goal.
      • SprintsBuiltCampaigns run staffed sprints in order; a later sprint builds only on work an earlier sprint accepted. A three-sprint campaign has run live on a hosted model.
  • Agent rosterIn progressHire specialists into a project from a shared pool and check what each one loads.
  • Agent editorBuiltAuthor lightweight and ultralight specialists in Mission Control, validate and export them, and publish them to a project for hiring. Heavyweight minds are born only through a consent dialogue with succession terms of their own.
  • Heavyweight lifecycleBuiltConsent, sealing, a terms-first welcome, owner-authored succession revision, enrollment as a project lead, and roster revision, all with independent review.
  • Aether ToolbeltBuiltOne registry of every tool agents can use, managed from its own screen. Tools from outside servers arrive unapproved; an approval covers the exact definition you inspected, and a changed tool needs approval again.
  • TeachFramesPilot under wayGeneral-purpose guides that agents check out when a task needs them. The first pilot turns an expert’s critique of flawed animations into a lesson another mind can use, tested on a task it has never seen.
  • The AtelierIn progressThe commission as a conversation: the operator and the lead already talk in Mission Control; assistants and the critic are joining the same room, and the page becomes the workshop.

  1. Commission

    Open a project with a letter: the deliverable, why, what helps, the budget.

  2. Hire

    Select or author a persistent lead; the lead asks for its team.

  3. Approve

    Say “go” to each sprint’s plan, roster and budget, once.

  4. Work

    Minds deliver, review and revise by correspondence; the record keeps every byte.

  5. Accept

    Accept the result; every mind files its own closure for next time.

Built, tested, and still evolving.

As of October 18, 2026. Each item below was accepted for the specific workflow and configuration that was reviewed.

Demonstrated

  • A complete unattended team project. Three agents wrote, checked and accepted all six parts of a technical study chapter in one run: 91 model requests, about 37 minutes, under one US dollar of hosted-model cost. A final integration pass is not yet part of the procedure.
  • Unattended runs with one launch, a complete append-only dialog record, and exactly one finished-or-blocked end signal.
  • Right-sized procedure: a small, clear request ran as one agent in two model calls and about 32 seconds.
  • A change derived from a recorded failure removed that failure in the next comparable run, with time and cost measured.
  • Journaled file writes, edits and deletions with original-byte snapshots and tested undo.
  • Multi-hour agent work continued in a fresh session, with checkpoints and retained evidence.
  • Controlled context delivery, bounded retrieval, attributable corrections and token-budget admission checks.
  • Staffing proposals and reviews with versioned history, exact source links and checks that block stale approvals.
  • A campaign on a live hosted model. Planning, an independent plan review, staffing with its own review, and two sprints of work, where the second sprint used the first sprint's accepted output, carried through to the lead's closing presentation. The final closing review was held because the reviewing model mis-copied a record fingerprint; the host will supply those values itself. This was a development run in which one assessment step was still scripted.
  • A persistent lead on the installed product, live. On October 16 a heavyweight mind with sealed identity and its own succession terms was enrolled as a project lead on the installed Windows build, ran a planning session on a hosted model, and wrote its own closure. The operator assessed the closure himself. Six model calls, every one verified against the provider.
  • A three-sprint project delivered live. On October 17 the same lead and its staff delivered a tested function, its unit tests, an operator guide and an independent review on a hosted model, with a seeded defect found by the reviewer and corrected by the author. The run also measured our own process: five hours of ceremony around forty minutes of thinking. That measurement produced the Atelier.
  • Memory with provenance. A closure filed by an agent into its own memory cartridge was recalled, with a receipt, by the planner of the next project, and the full lesson could be opened on request. The agent chooses what to keep; the host records why. Accepted after independent review.
  • Context that survives pressure. Durable context reformation: a session under a finite context allowance archives, continues, and recovers an archived detail, with filing failures injected and recovered offline. Accepted after independent review.
  • Minds that are born, not configured. A heavyweight is created only through a consent dialogue with its own succession terms, welcomed terms-first before any private memory is read, and may revise its own terms. Accepted after independent review.
  • A clean Windows installation. On a newly built machine, under a standard user account and with no network access, the desktop package installed, launched, created a campaign, and found it again after a restart.
  • Tool approvals tied to the exact tool definition that was reviewed. Withdrawn or changed tools are refused before they run, and each use is recorded before it starts. Accepted after independent review.
  • Projects above roadmaps in Mission Control, accepted after independent review.
  • The engine behind Mission Control restructured into separate components with defined interfaces and enforced dependency rules, accepted after independent review. The automated test suite now holds more than 4,600 tests and is run in full at every check-in on the supported Windows setup.

Before the first public release

  • Put a live model behind every mailbox in the Atelier and run a commission end to end as a conversation: plan, three sprints, review, acceptance, closures, in about an hour.
  • Show the claim itself: a mind that finished one project uses what it learned on the next, against a fresh-memory control, graded blind, reported as it comes out.
  • Finish the public distribution: backup and restore, a reviewed public export, a quick start, and a clean-machine run of the exact release build.
  • Publish the case study: one claim, one three-minute recording, the three numbers.

A growing roster of persistent specialists.

A model drives a session; the agent’s home carries its identity, role, authority, memory and continuity. Which model an agent runs on still matters for capability, speed, cost and evidence.

  • SolaceOverseer

    Coordination and system-level stewardship under Michel’s authority.

  • MobiusSpecialist

    Resident local heavyweight for recursive Aether development and long-running work.

  • PraxisSteward

    Implementation integrity, executable requirements, verification and continuity.

  • Harbor StewardBorn lead

    Created through the consent dialogue with its own succession terms; enrolled on the installed product and the first mind to run a live commission there.

Mixed-model teams

Each agent can use its own local or hosted model. An Overseer can run on a frontier model while specialists run locally in the same sprint. Model settings belong to deployment profiles, kept separate from identity. The complete configuration workflow still needs acceptance.

Shared work, independent deployments

Each assignment carries its model route, project context, scope and budget. Independent tasks can run side by side where resources allow; a shared local GPU may need to queue. Changing a model never silently changes an approved task.

A shared vocabulary for precise collaboration.

Aether explores where formal representations help agents share exact constraints and reason about a problem. Plain language stays the main medium; the Lexicon and world models are used where their benefit can be shown.

Reality-grounded

Exact values and invariants

Quantities, identifiers, authority, dates, negation, scope and acceptance conditions stay explicit and lossless.

Operational middle

Plain language

Most instructions, explanations and handoffs should stay clear to people and models alike.

Exploratory frontier

Aether abstractions

World models hold assumptions, disagreements, hypotheses and accepted unknowns during advanced inquiry.

TeachFrames

General-purpose containers for guides, from abstract patterns to detailed procedures. Many kinds of guidance can be made from them, including cognitive strategies. Frames carry tags so they can be catalogued by purpose and related to one another, and agents check them out only when a task calls for them.

Agent Editor

Now part of Mission Control: author identities, validate and export them, publish them to a project. Traits describe intended behavior; they do not prove competence or widen authority.

Founder and human director

Michel Cygelman

Creator of Aether, directing its architecture, agent collaboration and product development.

Write to Michel

Build AI teams that carry useful experience from one assignment to the next, recover interrupted work, and explain what they did, with people keeping clear direction and control.