A new supplier writes in; your Aivell sends the vendor pack, reads each returning document onboard — company details, tax ID, insurance, bank data — checks the set for completeness and consistency, chases what is missing, and registers the approved vendor in your database. Bank details always pass a human before anything is saved.
starts when: A new supplier sends their details to purchasing@
Onboarding a supplier sounds like one task and is actually eleven. Request the documents. Receive four of six. Notice the insurance certificate expired in March. Ask again. Receive a photo of a form, sideways. Check the IBAN against the letterhead. Type everything into the vendor master. Weeks pass between “we’d like to work with you” and the first order — and purchasing did nothing in that time but clerical relay.
The stakes are higher than tedium. The vendor master is where payment fraud walks in: a mistyped or maliciously “updated” IBAN sends real money to the wrong account, and the gap is usually discovered only when the genuine supplier asks where their payment went.
The guardrails do the quiet security work: the AI can create vendors but can never edit an existing one’s bank details, and its outbound mail is confined to the applicant and your purchasing team. Your onboarding procedure — documents, patience, thresholds — is the chore, written the way you would explain it to a new buyer on their first day.
No flowcharts, no code — a chore is just a message to your Aivell, in your own words. It shapes it into a solid, guarded procedure and follows it to the letter. This one:
when a new supplier writes in, send the vendor pack — everything we need, one list
as documents arrive, read each one — company details, tax id, insurance, iban
check the set is complete and the company name matches across all papers
missing or expired documents — reply listing exactly what is outstanding
still incomplete after a week, remind once; after three, tell purchasing
bank details never go in unchecked — Nadia confirms every iban before registration
register the approved vendor in the database and confirm to the supplier
note stalled or rejected applications for the monthly purchasing review
Connected with single-click passkeys — no copied tokens, no OAuth hell.
Every tool can be limited to exactly what this chore needs. For this one, you might set:
Two ways at once. The chore forces a pause — Nadia sees the submitted document and the extracted IBAN side by side and confirms with a tap before registration. And a guardrail makes existing vendors' bank details unwritable by the AI entirely, so a spoofed "we changed banks" email has no lever to pull.
Yes — the vision runs onboard the box, so a phone photo of an insurance certificate is read as easily as a clean PDF, and with the same privacy. If something is genuinely unreadable, the supplier is asked to resend, not guessed at.
Inside your office. Company registrations, tax IDs and bank documents are read and checked on your own hardware — which is exactly the privacy-by-design posture [GDPR Article 25 asks for](https://seresa.io/blog/eu-ai-act-for-marketers/gdpr-article-25-and-local-ai-why-on-premise-llm-inference-is-privacy-by-design).
The chore keeps the file open, reminds on your schedule, and after the patience you defined runs out, the application lands on the monthly review list with its history. Half-onboarded vendors stop haunting the purchasing inbox.
An on-premise AI colleague requests, tracks and chases client certificates ahead of the deadline — every send logged, nothing processed in the cloud.
Supplier invoices matched against purchase orders behind your firewall. Clean matches approve themselves for payment; only the mismatches reach a human.
An on-premise AI colleague reads each incoming contract, logs parties, value and renewal terms in a register — and books the deadlines on your calendar in time to act.
An on-premise AI colleague for manufacturing back offices: turns order emails into system entries, works supplier portals, watches stock, reads shop-floor photos.
Shipment status answered, PODs read and filed, carrier portals worked overnight — an AI colleague that keeps your rates behind your firewall.
Quote requests logged, site photos turned into reports, subcontractor certificates chased — an AI colleague that keeps your bids on site.
An Aivell is an on-premise AI colleague: a small box that plugs into your network, runs its own AI onboard and takes the repetitive work off your desk. Prompts, documents, data — nothing ever leaves your office. Unplug the internet: it keeps working.
You describe each task as a chore, in plain language. Your Aivell turns it into a solid, guarded procedure and runs it in the background — with guardrails on every tool, a complete audit trail, and approvals in your hand when you want them. One-time setup, fixed monthly fee. No tokens, no overages.
onboard inference · air-gapped · fixed monthly fee