Skip to content
BeryxTech

Lighthouse Projects · HR

AI helpdesk for HR and IT

The usual questions answered by the agent, and the sensitive ones by people.

Problem

High volume of repetitive tickets on vacation, benefits, access, policies. Long response times, HR and IT teams with no room for strategic initiatives.

Solution

AI agent in chat (Teams/Slack/portal), with KB tied to policies and FAQs. HCM integration for queries (vacation balance) and simple actions (open ticket, password reset). Smart escalation for sensitive topics.

KPIs

  • % of queries auto-resolved
  • Average response time
  • Hours/month freed in HR and IT
  • Employee satisfaction (eNPS)

How it is measured

Every KPI with a baseline and a target.

The numbers for this project come from your data. We measure the baseline during framing, agree the target with you before building and, in production, measure the same KPIs the same way.

  1. 01BaselineMeasured during framing, with the data of the process as it runs today.
  2. 02TargetAgreed with the case owner before building, starting from the baseline.
  3. 03In productionThe same KPIs, measured the same way, compared with the baseline.
% of queries auto-resolved
Conversations closed without handing over to a person or opening a ticket, divided by all conversations.
Average response time
From the request to the first useful answer, from the chat and ticketing logs.
Hours/month freed in HR and IT
Repetitive tickets that no longer reach the teams, multiplied by the average handling time measured before.
Employee satisfaction (eNPS)
A short survey at the end of the conversation and the company eNPS, before and after.

Before starting

Prerequisites and risks.

Prerequisites

  • Written, current HR and IT policies and FAQs, with an owner to maintain them.
  • Access to the HCM for lookups, such as leave balance, and to the ticketing tool.
  • The list of sensitive topics that always go to a person, defined by HR.

Risks and how we handle them

  • Answers outside the policies

    The agent answers only from approved policies and cites the source. With no answer in the knowledge base, it routes to HR or IT.

  • Employees' personal data

    The agent only sees the data of the person asking, using the HCM roles. Health, conflicts and performance reviews go to a person.

  • Transparency and AI Act limits

    Employees know they are talking to an AI agent, as the AI Act requires (Article 50(1)). An agent that decides on working conditions or evaluates performance becomes high-risk (Annex III, point 4(b)) and is outside this project.

    Sources

Other projects · HR

Want to see this in your operation?

90 minutes with you and your team. We leave with 3 concrete opportunities and estimated payback.

Book diagnosticAll Lighthouse Projects