# Email — architecture finale + liste des emails du staff + logins username

> ⚠️ **Version finale à envoyer : `email-roma.txt`** (v2 retravaillée par Bastien :
> voix « je », destinataires Roma + Narin, détail concret du routing, cadrage
> « step one / what comes next », Ollama coming soon, format texte brut).
> Le présent fichier = v1 conservée pour l'historique.
> Pièces jointes : `diagrams/adr-0002-ai-assist.png`,
> `diagrams/final-architecture.png`, `staff-email-map.csv`. Réponse attendue → `ANSWER.md`.

---

**Subject:** Mail system: final architecture, and two things we need from you

---

Hi Roma,

Quick update on the pilot, the architecture we've settled on, and two pieces of
information we need from your side to move forward.

## 1. Where we stand

The core is working end-to-end on the test domain: incoming emails are sorted
automatically into the right team queues (we ran a realistic 13-email test round, and
all 13 landed correctly), your hybrid queue layout is live, and the AI drafting
assistant is validated inside the helpdesk. The first attached diagram
(**adr-0002-ai-assist.png**) shows how drafting works today: the assistant proposes a
first draft, then a person always reviews, edits and sends. **Nothing ever goes out
automatically.** That rule is built into the system, not just a policy.

## 2. The final architecture, and why we built it this way

The second diagram (**final-architecture.png**) shows the full structure. The idea is
simple: **we assemble proven open-source products as modules, and we only write one
small piece of software ourselves.**

- **Zammad**, the helpdesk module. Receives all mailboxes, sorts requests into team
  queues, tracks status, and is where your team replies from.
- **Open WebUI**, the AI workspace module. A chat interface where your team can ask
  the assistant things like "what's urgent in the quote queue today?", and where the
  company knowledge lives (MOQ policy, sampling fees, business terms, past quotes).
- **Claude (Anthropic)**, the AI model behind the assistant.
- **The connector**, the only piece we actually code: a small, auditable bridge that
  lets the assistant read tickets and place **draft** replies inside the helpdesk. By
  design it has no ability to send email. Only humans send.

Why this matters for you:

- **Each module is maintained by its own team.** Zammad and Open WebUI are widely-used
  open-source products with their own developers, security updates and documentation.
  We don't modify them, so updates stay safe and nobody inherits a pile of custom
  hacks buried inside each tool.
- **Maintenance stays realistic.** One small connector to maintain instead of custom
  code scattered across every product. If something needs fixing, there is exactly one
  place to look.
- **It's future-proof.** Other modules can be added later (accounting, CRM, product
  catalog and so on) the same way: plugged in through the same pattern, not welded in.

What the AI workspace adds on top of what the pilot already does:

| Already working today | What the AI workspace module adds |
|---|---|
| Automatic sorting of incoming emails into team queues | Ask questions across all tickets ("anything urgent in Triage?") |
| One-shot AI draft inside the reply window | Full back-and-forth conversation about a ticket before drafting |
| | Drafts informed by company knowledge: MOQ, sampling fees, terms, past quotes |
| | Complete draft placed directly into the ticket, ready for human review |

## 3. What we need from you

**(a) The complete email map of Advance Manufacturers.** I attached a short
spreadsheet (**staff-email-map.csv**, opens in Excel) to make this easy. Please fill
in one line per person for everyone who sends or receives work email: full name, role
or department, and the email address they use. If someone uses several addresses,
add one line per address.

In the same file, please also list every **shared or company mailbox** that exists
(info@, sales@, accounting@ and similar), even old or barely-used ones. We need the
full picture to map mailboxes to queues and decide what gets plugged in.

**(b) Usernames for logging in.** We've confirmed the helpdesk supports simple
**usernames** for login instead of email addresses (for example just `noc`; this is
internal only and doesn't change anyone's email). Please confirm you're happy with
that, and use the "Preferred username" column in the spreadsheet to tell us what each
person wants. If left empty, we'll default to first names.

**(c) Your current email signatures.** So replies from the new system look exactly
like your emails today, please send us the signatures currently in use: the company
signature template, and each person's variant if they differ (name, title, phone,
logo and so on). The easiest way is to forward us one recently sent email per person;
the signature at the bottom is all we need.

**(d) Your reply templates, per department.** If your teams have standard replies
they reuse (quote replies, sampling fee explanations, payment terms, shipping
updates, complaint handling and so on), please send them to us, organized by
department. And if possible, include a few real case studies: complete email
exchanges from first customer contact to the final reply (customer names can be
removed). These are what we'll use to teach the assistant your tone, your terms and
your way of answering, so the drafts it prepares sound like Advance and not like a
generic robot.

Nothing else is needed from your side for now. Once we have the list, the signatures
and the templates back, we'll prepare the accounts, the signatures and the final
mailbox mapping for your validation.

Best regards,
Bastien

---

**Attachments:**

| File | What it is |
|---|---|
| `adr-0002-ai-assist.png` | How AI drafting works today: draft proposed, human reviews and sends, never automatic |
| `final-architecture.png` | The final module structure: open-source modules plus the one connector we code |
| `staff-email-map.csv` | Template to fill in: people, roles, email addresses, shared mailboxes, preferred usernames |