Fewer routine questions and less admin, more time for the people who need a professional.
An AI agent answers routine questions from verified information, prepares intakes and takes over admin. It makes no diagnosis and no treatment decision. With sensitive questions it hands over to a professional. Veilig Thuis held more than 15,000 conversations this way in four months.
The work around care, so professionals have time left for the care itself.
Answers to recurring questions day and night, based only on your own verified information.
View Customer Questions Agent →The agent asks the questions from your intake protocol, checks whether the registration is complete and passes it to the right team member with a summary.
View Intake & Triage Agent →Schedule, reschedule and confirm appointments, without phone tag.
View Appointments Agent →Answer the phone when the team cannot, and handle recurring questions directly.
View Voice Agent →Read and sort emails and draft replies. A team member checks and sends.
View Mailbox Agent →Bring figures from several systems together in the report that now takes a day.
View Reporting Agent →These assistants provide information from verified sources. They show how AI works in a context where care comes first.
No diagnosis or treatment advice: software that does give it can fall under the medical device rules. We check this upfront
Emergency healthcare triage is high-risk under the AI Act. In an emergency the agent refers people onward
Health data is special category data under the GDPR. The DPIA is included with the agent by default
Answers come only from verified sources. If it is not in there, the agent says so and hands over
The agent always says it is AI, and a person remains ultimately responsible
The same three packages as for every agent. We decide which one fits together in the process call.
Question not listed? Ask it in a process call
No. The agent provides information from your verified sources and refers people onward with medical or sensitive questions. The decision stays with the professional.
Hosting and data are in the EU, every action is logged and the DPIA is included by default. We record upfront, per process, which data the agent may use. An assistant can also work anonymously, as at Veilig Thuis.
We decide that together upfront. The agent recognises the signals you define and refers people straight to the right number or team member. It does not try to resolve such a situation itself.
If the system has an API or a standard message format, we connect it. We check this upfront in the process call.
In a 30-minute process call we look at which work an agent can safely take over, and which it cannot.