AI you can actually let into the building.
A chatbot that says something wrong is awkward. An AI employee that does something wrong is a problem. So we built Winsen so the wrong thing is hard to do, and easy to catch.
Permission. Approval. Audit.
The three questions every security team asks, answered by design.
Can it even see that?
Walle and the AI employees inherit the exact permissions of the person they work for. If you can't see a doc, neither can your AI. No new access surface, no data leaking across teams.
- →Inherits your existing roles and scopes
- →Respects every source system's permissions
- →No "AI admin" backdoor to your data
Should it really do that?
Consequential actions wait for a human. Sending money, emailing a customer, deleting anything, those stop for approval, every time, until you say otherwise.
- →Money, outreach, deletion always gated
- →You choose what becomes autonomous
- →Every gate is configurable per role
What exactly did it do?
Every action an AI employee takes is logged with its reasoning, the data it touched, and the result. Not a black box. A glass one.
- →Every action logged with its reasoning
- →Filter by employee, action, or outcome
- →Export for your compliance team
“What exactly did it do?” Click and find out.
Every fact carries its source and how sure the system is. Same for every action. This is the actual audit view, from a live workspace.
Your data stays yours.
The brain is customer-owned and exportable. We don't train on it. We don't sell it. You can take it and leave whenever you want.
- No training on your data, ever.
- Export the full brain on demand.
- Delete everything, for real.
Compliance, built in. Not bolted on.
We built Winsen for SOC 2, HIPAA, and GDPR from the ground up, so the controls are live in the product today. The formal certifications are admin in motion, and we'll tell you exactly where each one stands.
We'd rather tell you exactly where we are than imply we're further along. Ask for the current report or a BAA.
No black-box runtime. We show our work.
The engine under Winsen is open and inspectable, not a proprietary mystery you have to trust.
Platos
The runtime your AI employees think on is open source. Not a proprietary black box you take on faith. Read it, audit it, self-host it if you want to.
platos.dev →Trigger.dev
Every task and workflow runs on Trigger.dev: durable, observable, retryable. Long jobs survive restarts, and every run leaves a trace you can inspect.
trigger.dev →The honest answers.
No dodging, no contact-sales-to-find-out.
Is my data used to train models?+
What happens if the AI does something wrong?+
Are you SOC 2 and HIPAA certified?+
What runs Winsen under the hood?+
Where is my data hosted?+
Can I delete everything?+
Who can see what the AI sees?+
The boring stuff, done right. So the exciting stuff is safe.
Want the deep version? Read the security docs or talk to our team.
