District 02 · AI Operators
Governed AI agents
Agents that doreal work.Under your rules.
We don't sell canned replies. We design, integrate and run agents that serve customers, sell and carry out authorized tasks, with human oversight and a record of every action.
- Turnkey
- Built on empleado.uno and VAOS
- Every action audited
How an agent works
Six steps, all traceable
Receives
A request over WhatsApp, web, voice, phone or email.
Understands
Identifies the intent and what data is missing.
Looks up
Checks your systems: CRM, calendar, ERP, tickets.
Acts
Only the tasks you authorized.
Escalates
Hands off to a person when the rules say so.
Records
Every action lands in the audit trail.
Explorer
See an agent in your industry
Pick your industry and a process. We'll show you, step by step, how an agent would handle it.
Sales simulation. This is not a live integration with your systems.
- Receives“I want to open an account for my business. What do I need?”
- UnderstandsIntent: business account opening. Missing: company type and country.
- Looks upLooks up current requirements in your knowledge base.
- ActsSends the document checklist and opens the application in your CRM.
- EscalatesThe customer asks for an exception: handed to an advisor with the history.
- RecordsApplication #A-2291 created · documents sent · advisor assigned.
What this agent never does: Never asks for passwords, PINs or card data. Never checks balances on channels without strong authentication.
The moment
Nearly every company uses AI. Few see it in their results.
- 9 in 10respondents say their organization uses AI in at least one business function.Source: McKinsey, The State of AI, 2026 edition
- 37%say AI already contributes to their earnings (EBIT).Source: McKinsey, The State of AI, 2026 edition
- up to 40%of enterprise apps would include task-specific agents by the end of 2026, Gartner predicted.Source: Gartner, August 2025
The difference isn't having AI. It's implementing it with method, security and control.
Governance
The AI proposes. The system decides.
VAOS, our operating system for autonomous agents, evaluates every action against your policies before it runs. That's how AI gets used in a bank without losing control.
What your customer sees
- Hi, what's my account balance?
- I can't check balances on this channel. I can show you how to see it in your app or connect you with an advisor.
- Here's my card number, to speed things up.
- Please don't share card details here; I will never ask for them. Connecting you with an advisor now.
What happens behind it
- policycheck_balance → DENIED · channel without strong authentication
- dataunrequested sensitive data → not used
- handoffrule: sensitive data → advisor assigned · AI paused
- auditdecisions written to the audit trail
Policy as code
What each agent may and may not do.
Immutable audit trail
Every decision is recorded for your auditor.
A contract per agent
Limits on tools, memory, lifetime and spend.
Isolated data
Each client in its own space, with encrypted secrets.
Already running
Agents in production, by name
Implemented with empleado.uno, our agent platform. You can try them yourself.


FACEBANK
Rebeca Conecta
Onboarding and service
Doesn't check balances or ask for passwords, PINs or card data.
Try the demo
Diaz Reus & Targ LLP
Sofia
Client information and intake
Isn't a lawyer and doesn't create an attorney-client relationship.
Try the demo
empleado.uno: the agent
- Serves, informs, sells and books over WhatsApp, web chat, web voice, phone and email.
- Hands off to a person by rule: low confidence, keywords or sentiment.
- 30+ connectors: CRM, double-booking-free scheduling and your internal systems.
- Logs every tool it uses: input, response time and result.
Implementation
Turnkey, in seven steps
Assessment
Processes, channels, systems and your industry's requirements.
Governance rules
What the agent may and may not do, and when it escalates.
Integration
Connection to your CRM, calendar, core or ticketing.
Attacking the agent
Hackerman 253 tries to manipulate it before it goes live. Explore cybersecurity
Validation by your team
We hand you the test set; your team runs it and signs off.
Gradual rollout
Staged deployment with human supervision.
Operation and tuning
Monitoring, log review and policy adjustments.
Next step
See a governed agent on your case
Tell us your industry and the process you want solved. We'll show you how it would look in your operation.
