Skip to content

Compliance comes first.

Putting AI in front of your team or your customers should not start a review you dread. These are the choices already made in Lindy Logic, in plain words, for whoever has to sign it off.

The old town of Tallinn, Estonia, seen from a viewing platform.

Your data

Conversations belong to your account
Export them, keep them, or delete them.
Files stay private
Documents you add and files the assistant writes sit in private storage and are handed out through short-lived signed links.
Each team is kept apart
A team's conversations, tables and files are separated from every other team's.

Your website

The embed checks where it runs
An assistant embedded on your site is checked against the site it is allowed to run on.
No third-party fonts
Fonts are self-hosted, so loading the page does not send your visitors' IP addresses to a font host.
A content security policy
The embed runs under one, which limits what the page is allowed to load.

Signing in

No passwords
People sign in with a link sent to their email. There is no password to forget, reuse or guess.

AI, said plainly

Replies are labelled
Each assistant reply is marked as written by AI, and every conversation says that replies can contain mistakes.
You choose the model
OpenAI, Anthropic, Mistral or Google Gemini, for the whole team or for one guide.
Code keeps data closer
Because the assistant writes code that runs on our servers, less of your data needs to go to the model provider.

Who you are dealing with

Lindy Logic is built and run by Radical Software OÜ in Tallinn, Estonia: one company and one contract, under European rules. Read the privacy policy and the terms of service.

No system is foolproof, this one included. If your review needs something that is not on this page, ask and we will answer in writing.

Ready to accelerate with confidence?

Bring whoever signs off your tools to the call. Half an hour is usually enough.