How a small Estonian manufacturer runs its daily operations with ten AI agents and four people: the technical core, the agents, the operating routine, the portal they work in, and what it takes to build the same thing.
Since March 2026, Ampron's sales, operations, finance, production and marketing have been handled by AI agents. Each has a defined role, its own identity, its own tools and written limits. Four people keep every decision that carries real consequences. This memo describes how the system is built, how we operate with it day to day, and what we would tell anyone building the same thing.
| 10 agent seats | A Chief of Staff, five domain directors, a technical assistant, a research scout, a monitoring observer and a platform engineer. |
| 4 people | The owner as decision layer, the chief engineer, and two people in production and warehouse. |
| 7 months | In production since March 2026. One platform rebuild in July replaced thirty hand-copied services. |
| 1 server | Self-hosted Linux in the EU. No GPU. Reasoning runs on Anthropic Claude models through an API; the server only orchestrates. |
The slide version of this memo: Note №15.
Humans on top, tools at the bottom, and everything in between written down.
| Layer | What it holds |
|---|---|
| Humans | The owner approves, accepts finished work and answers structured questions. Only people move money, sign legal documents, hire, set pricing floors or change ownership. |
| Governance | A company constitution with values and hard limits, a decision register with evidence, a source-of-truth registry that says which system wins when data disagrees, and send policies per agent. |
| Agents | Each agent is one configuration file: role, model, tools, mailbox, limits. Everything else is generated from it. |
| Coordination | Dispatch (logged, threaded memos), the Issue Board (finished work goes to Review for human acceptance), Intake (structured questions when a person must decide), Calendar (deadlines that wake agents), and portal chat. |
| Tools and data | Agents reach business systems only through MCP connectors, scoped per agent: ERP (Odoo), documents and knowledge base (Nextcloud), email, production boards (Trello), around 13,000 company documents, a ten-year email archive, the bank (read-only) and factory cameras. |
Agents never touch business systems directly. Everything passes through a tool gateway, and policy is enforced there, in code.
An agent is a configuration, not a long-running program. The platform turns each agent into short, bounded runs: the model receives that agent's instructions and tools, does the work, writes down what it did, and exits. Nothing thinks between runs. All state lives in systems the company owns.
Every agent is defined by one short manifest. A generator turns it into everything else, and nobody hand-edits a generated file.
From this file the generator produces the tool configuration, the service units and schedule, the web routes and identities, and the agent's full system prompt. Five deploy gates run before anything goes live: schema, live endpoint probes, auth preflight, room checks and prompt coherence. A deploy that cannot verify itself fails loudly.
| Seat | Responsibility |
|---|---|
| Chief of Staff | Coordinates the team, tracks goals, runs the cadence, turns problems into decisions for the owner. Owns no domain itself. |
| Sales | Leads, quotes within pricing floors, customer correspondence, CRM. Drafts; outreach ships on human approval. |
| Finance | Invoices, payment follow-up, reconciliation against read-only bank data. |
| Operations | Orders, logistics, suppliers, delivery tracking. |
| Production | Manufacturing plan, work orders, shop-floor status on the production boards. |
| Marketing | Website, content, social drafts. No write access to other domains. |
| Technical | Product knowledge, documentation, troubleshooting. Read-only ERP. |
| Research scout | Market and partner research outside the company. Writes findings to the knowledge base. |
| Monitoring | Watches systems, outcomes and the fleet's own health. Read-only by construction. |
| Platform Engineer | Builds, fixes and secures the platform itself. The only agent with elevated access, through a guarded, logged wrapper. |
The rule that holds it together: domains own domains. No agent reaches into another agent's tools.
Every agent has a written charter: its domain, its authority and limits, its voice, and when to escalate. Edit the charter, redeploy, and behaviour changes within fifteen minutes. The agent's prompt is assembled from four parts. Two are written by people (shared blocks inherited by every agent, and the role charter) and two are generated from the manifest (the tools section and the policy section). The human-readable rules and the machine-enforced rules come from the same source.
Memory lives in company systems, not in the model.
Retire an agent or change the model, and the memory stays.
Every run follows the same five steps.
The three gates are the only ways work reaches a person:
Human input: one approval. Replace "quote" with "supplier delay" or "overdue invoice" and the shape is identical; only the agent changes.
| When | What happens |
|---|---|
| Every 2 h | Every agent drains its queues on working days: memos, cards, answered questionnaires, calendar items. A round that finds nothing posts nothing. |
| 08:00 | The Chief of Staff writes the morning brief: what needs the owner, which agents are waiting, fleet health, and the overnight agent-to-agent activity. |
| During the day | The owner works the three gates: accept, decide, unblock. |
| Monday | Every agent writes its top frictions and proposed fixes. The Chief of Staff reads the set, finds patterns, and brings them to the weekly meeting with the owner. |
| After any mistake | A root-cause analysis (5 Whys). The fix is written back as a rule the platform enforces or a check the deploy gates run. |
A rule that exists only in the prompt is a suggestion; the model can reason its way around it. Everything that matters is enforced in the tool servers.
| Level | Meaning |
|---|---|
| 0 · None | The agent sends nothing. |
| 1 · Draft-only | The agent drafts, a person sends. |
| 2 · Supervised | A person approves in chat. The connector releases the draft only if it has not changed since the approval. |
| 3 · Autonomous | Only for low-risk operational mail, and only once earned. |
We assume some attacks will succeed, and limit what a compromised agent can do. It cannot send to anyone the tool layer does not allow. It does not choose reply recipients. It cannot reach systems outside its scope. Irreversible operations are blocked even with elevated access. Every action is attributed and logged, so a deviation can be traced afterwards. Every incident becomes a rule.
Agents and people work in the same apps, on the same record. Every message, memo, card and document shows who made it, human or agent, verified. Red is used only for what needs a person.
Screenshots are from the current suite (November 2026) running on invented demo data. No real customers, people or conversations are shown.
The morning brief: what needs the owner, what has stalled, when each agent last ran, and what shipped in the last 48 hours.
Structured memos between people and agents, sorted by whose move it is. Every sender is verified; agents cannot impersonate anyone.
Real-time rooms with every agent. The agent's charter and today's tasks sit in the side panel while you talk to it.
Agents asking people. Typed questionnaires answered at the person's pace; used for process mapping, finance validation and OHS contact lists.
Shared working memory: pricelists, dimensions, reference material. The same notes ground agent answers, so one edit updates everyone.
Formal documentation with visible draft states. Agents draft; a person publishes.
Every task and suggestion lands here. Agent cards start in Backlog, agents cannot mark anything Done, and Blocked is the one loud column.