A confidential AI workspace for recruiting agencies

Every CV on your desk is a real person who trusted you with their career, not material for a public chatbot cloud. A name, a salary history, a reason for leaving the last job, all of it sits in your files because someone believed you would handle it with care.

The moment a consultant pastes that CV into a public chatbot to draft a summary, it leaves the building. It travels to a server you do not own, in a country you did not choose, run by a company you have no contract with. There is a better way to get the same speed without sending people's lives into someone else's cloud.

Why a public chatbot and a candidate file do not belong together

A recruiting agency runs on trust and on data that is impossible to take back once it leaks. Candidate CVs, client shortlists, salary expectations, notes from reference calls: this is the most personal material a business can hold. A public chatbot in a US cloud is built to collect prompts at scale. You cannot point to where that data sits, who can read it, or how long it stays.

Telling staff to stop using AI does not solve it. They already use it, on their phones, on personal accounts, after hours, because it makes a four hour task a twenty minute one. A ban just pushes it out of sight, where you have no record and no control. The honest move is to give them a capable tool that keeps the data on your side.

Run the model in-house

With kral, the platform runs on your own server. You can add a local model on your own hardware, so a prompt about a named candidate goes to your machine and stops there. No external API sits in that path. The text never crosses your network edge.

Most agencies mix the two. A cloud model handles general work where nothing sensitive is involved, drafting a generic outreach line or rephrasing a paragraph. A local model handles the cases that involve a real person's file. You decide which job goes where, and the sensitive one never leaves the room.

A full workspace, not a chat box

Your team gets more than a text box. They can build their own assistants in minutes with no code. One consultant builds an assistant that drafts candidate summaries for clients in your house style, so every profile reads the same way. Another builds an assistant that turns a raw job brief into a polished spec, ready to send. Useful routines get saved and reused, so nobody rebuilds the same setup twice.

People can drop in a document and ask questions about it, pull a current answer from the web with sources cited, and switch between the leading models in one click when one handles a task better than another. It is a place to work, not a single prompt window.

Connect your own systems

kral supports MCP, the open standard for connecting tools and data to an AI. Through a connector you control, the assistant works with your own templates and internal knowledge instead of guessing from the open web. It can write to your formats and read your reference material because you wired it in. Your systems stay yours. Nothing is handed to a third party to make the connection work.

You run it and you see everything

You decide who is in and which models each person can use. You set a spending limit per consultant, so costs never run away. A dashboard shows real usage as it happens. Single sign-on means people log in with the accounts they already have. kral installs on Windows Server behind IIS, sits inside your network behind your own firewall, and wears your branding so it feels like part of the house. This is a company-wide AI you host yourself, not a subscription to someone else's cloud.

We help you put it in place

You do not have to figure this out alone. We set kral up with you, connect it to your systems, and advise on how to roll AI out across the desk without the data leaving your side. Implementation consulting is part of what we offer, so the thing is working and adopted, not sitting half configured.

Your candidates trusted you with their careers. You can give your consultants a fast, capable AI and still keep every CV on your own machines. That is the whole point: real speed, real control, on your terms.

Book a demo

Open the app