ZaffreZaffre Axon
Policy answers at 2 a.m., actions included

AI HR assistant

An in-app assistant that answers from YOUR company’s policies, shows only the asking employee’s own data, and can file requests on their behalf — running on company-controlled infrastructure.

The challenge

Half of HR’s day is answering the same questions: how many leaves do I have, what is the reimbursement policy, when does my probation end. The other half is processing the requests those answers lead to.

The Zaffre solution

Ask-AI-HR answers from the company’s own policy knowledge base, scoped strictly to the asking employee’s data — and when the answer leads to an action, it can submit the request right there, through the same approval chains as always.

Ask-AI-HR sits inside the workspace and answers what employees would otherwise ask HR: balances, policies, entitlements, procedures. Answers come from the company’s own knowledge base of policies, so they are correct for THIS company, not generically plausible.

Scope is enforced, not promised: the assistant reads only the data the asking employee is allowed to see — their own leave balance, their own payslips — with the same role-based permissions the rest of the suite enforces.

Where other chatbots end with advice, this one acts: an employee can ask it to apply for tomorrow’s leave and the request files into the normal approval chain immediately. HR gets fewer interruptions; employees get instant answers and instant filings.

Built-in safeguards against fake attendance

Your policies, not the internet

Answers ground in the HR policies your company loads into the knowledge base — updated instantly when policies change.

Own-data-only

The assistant sees the asking employee’s own balances and records — never a colleague’s. Role permissions apply to AI exactly as to screens.

Actions with approvals

It can file leave and other HR requests on the employee’s behalf — the requests enter the normal approval chains, nothing bypasses them.

Company-controlled AI

The assistant runs on infrastructure the company controls. Your HR data is not used to train external models.

How it works

  1. 1

    Load your policies

    HR maintains the knowledge base; updates apply to answers instantly.

  2. 2

    Employees ask

    Questions in plain language, answered from company policy and the asker’s own records.

  3. 3

    Assistant acts

    On request, it files the corresponding HR request through the normal approval chain.

  4. 4

    HR supervises

    Everything the assistant files is a normal, auditable request — visible and reversible like any other.

Frequently asked questions

Where do the assistant’s answers come from?
From your company’s own policy knowledge base plus the asking employee’s own records — not from generic internet knowledge.
Can it see other employees’ data?
No. Access is scoped to the asking employee’s own data under the same role-based permissions as the rest of the platform.
Can it actually submit requests?
Yes — common HR requests (leave and others) can be filed by the assistant on the employee’s behalf, entering the standard approval chains.
Is our data used to train AI models?
No. The assistant runs on company-controlled infrastructure and your data is not used to train external models.

See ai hr assistant in action

Book a demo