DAN AI — Getting Started
This page walks you through creating a DAN AI account, the layout of the dashboard, and choosing the best path to start extracting documents.
Sign up and sign in
- Visit dan.sdlccorp.com and click Sign Up.
- Use your work email and create a strong password, or sign in with Google.
- Confirm your email if prompted, then sign in.
You land on the Dashboard, which is also where you'll spend most of your time during normal operation.
Dashboard tour
The left sidebar contains the main sections you'll use:
| Section | What it does |
|---|---|
| Dashboard | KPIs and the most recent documents at a glance. |
| Documents | The full list of extractions, filterable by state, document type, source, sender, date and confidence. Also where you click New Extraction. |
| Inboxes | Connected Gmail and Outlook accounts, their rulesets, sync status and last-error info. |
| Developer | API keys, webhooks, and an activity log of recent API calls. |
| Settings | AI providers, default prompts, account-wide field defaults, and team management (admins only). |
Pick an ingestion path
Decide how documents will flow into DAN AI. You can mix and match — many teams use the dashboard for ad-hoc work, inbox sync for recurring vendor mail, and the API for downstream automation.
| Method | Purpose | Best for |
|---|---|---|
| Dashboard upload | Drop a single file from the browser through New Extraction, configure fields, and extract on demand. | Operations staff processing files one at a time. |
| Email inbox sync (Gmail / Outlook) | Auto-extract attachments from a connected mailbox on a schedule, with a rule engine for filtering. | Recurring invoice or receipt mail from known senders. |
| REST API | Send a file programmatically and receive structured JSON in return. Sync by default; queue an async job for long-running extractions. | Developers integrating DAN AI into their own apps. |
Continue with the walkthrough for the path you picked:
Recommended first run
Even if you're going to automate, do at least one dashboard upload first. It's the fastest way to see how DAN AI behaves on your own files, which document type fits best, and what confidence scores you can expect — knowledge you'll need before you wire up an inbox rule or write API client code.
Account-wide settings worth setting early
These live under Settings and are admin-only. Configuring them once saves time across every future extraction:
- AI provider chain — pick the primary provider (Ollama is the default) and the fallback order.
- Default fields per document type — if you always want the same 5 fields from invoices, set them here so every extraction returns just those.
- Custom prompts — override the built-in prompt for any document type. Useful when your invoices have non-standard headings or tax columns.
Next steps
Pick your ingestion method and follow the corresponding walkthrough, or read the Supported Document Types reference to see what DAN AI ships with out of the box.