Employees ask about leave, expenses, benefits or working hours in the chat widget; the Aivell answers by quoting the current handbook, section and all, in the asker's language. Personal cases go straight to HR, unanswered. Questions are processed on the box, and the weekly summary shows topics, never names.
starts when: An employee asks a policy question in the chat widget
“How many days of parental leave do I get?” has an exact answer, written in a handbook your company already paid to produce. Yet it gets asked by email, one employee at a time, and answered by hand, one HR person at a time. Email already eats 28% of the average workweek; a decent slice of HR’s share is questions the handbook answers verbatim.
The quieter cost is the questions that never get asked. Nobody wants to reveal to a colleague — even a friendly one in HR — that they are checking the sick-leave policy or what happens to their pension if they resign. So people guess, or ask the person at the next desk, and the guess becomes office folklore.
It behaves like the most patient person in HR: word-perfect on policy, discreet by construction, and honest enough to say “that’s not written down yet”.
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:
answer questions about leave, expenses, benefits and working hours
quote the current handbook and name the section
reply in the language the question was asked in
parental leave or long illness? give the policy, then loop in Sara
anything personal — salary, a dispute, a complaint — goes to Sara unanswered
if the handbook is silent, say so and log the gap
never record who asked what; the summary is topics only
friday afternoon, send Sara the week's topics and handbook gaps
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:
No — and that is deliberate. The chore forbids recording identities, and the weekly summary Sara receives lists topics and counts, not people. Employees ask the awkward questions precisely because nobody is watching.
No. The question, the search and the answer all happen on a box in your own server room. For questions like "what happens to my bonus during sick leave", that is the difference between a tool people use and one they avoid.
The Aivell says so instead of guessing — a guardrail pins it to the published handbook. The gap is logged, and Friday's summary tells HR exactly which policy pages are missing.
The chore draws the line — a salary question, a dispute, a complaint gets a polite handoff to Sara, unanswered. Policy is for the assistant; people are for people.
Every manual, contract and memo indexed on a box in your office. Your team asks in plain language and gets cited answers — nothing leaves the building.
Joiner and leaver checklists run themselves — IT notified, sessions booked, progress tracked — by an AI colleague that keeps HR data inside the office.
New documents arriving by email or landing on the share are read, indexed and made searchable automatically — on a box inside your own network.
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