KESHER: the same AI brain on every channel, and a human when needed
KESHER is Zooz-tech's AI platform for resident and customer service, built on Azure AI Foundry and Azure AI Search. The bot answers only from approved sources, verifies the resident's identity before any personal action, and hands off to an agent with the full context. Every conversation is logged in the local authority's CRM.
Generative UI: the bot builds the interface during the conversation
Buttons, cards and forms built inside the conversation, an honest answer when there is no source, and a handoff to an agent in one click.
Residents don't need to know where to turn
KESHER is set up on top of the existing website: our engine crawls and indexes the local authority's website, documents and content management system, and the bot takes on the authority's design and a voice, quickly and inside a secure Azure infrastructure.
Municipal website
Chat with cards, forms and buttons built to fit the answer.
The same capabilities, in short lines and buttons.
Voice on the website
Talk to the bot in Hebrew, English or Russian, including filling in forms by voice, while the screen shows the results in parallel.
Phone bot
Voice answering for inbound calls to the contact center, and outbound calls: reminders, request closure updates, callbacks and surveys.
The personal area, inside the conversation
In the conversation, residents get exactly what they would get in the personal area of the website, under the same rules and with the same wording, no less and no more.
Source-based answers
Only from indexed content: the municipal website, documents and the content management system. No authoritative source? The bot says it will not guess and offers an agent or a callback.
Identity verification before action
A one-time code checked against the CRM for any personal information. General information is open to everyone.
Personal information
Property tax (arnona), education payments, parking tickets, veterinary services, appointments and request status.
Opening a service request
"There's a broken streetlight near my house" becomes a classified request with an address, for the resident's approval, and is routed to the municipal 106 hotline.
A form inside the conversation
Field by field, with validation of ID number, phone, email and date, in text and by voice.
An agent at any moment
At any stage, residents can ask for an agent and choose a contact center. The conversation is handed over with the full context.
The agent joins the conversation with the full context
Live board
Conversations by channel, topic and agent, and SLA metrics in real time.
Agent Copilot
A suggested reply for every request, and a measure of how many suggestions were sent without editing.
Anomaly investigation
A spike in requests on a topic? In one click, the AI explains what happened, likely causes and recommendations.
Hazard map
Requests by street, with the leading topics, without house numbers.
Improvement loop
Questions the bot could not answer become tasks for the content team.
Outbound campaigns
Upload a recipient file, and the bot contacts them.
Answers from sources, protects privacy, and always leaves a path to a human
Grounding gates
An answer without a source citation, or with weak grounding, is blocked before it is sent.
Transparency
The bot identifies itself as an automated system and states that the conversation is recorded.
Privacy
Personal information only after identity verification, and identifying details are not sent for translation. The design includes compliance with Amendment 13 to Israel's Privacy Protection Law.
Accessibility
Full keyboard navigation, screen reader support and real buttons for quick replies.
Data residency
Static data in Azure Israel Central, AI processing in Sweden Central.
Active pilot at Rishon LeZion Municipality
A bot on the website, on WhatsApp and in the contact center console, connected to the municipality's Dynamics 365.
A local authority that wants this kind of service for residents?
We will show you KESHER running on your authority's own content.