Five open agents, one per area, working on the systems of a fictional company. No sign-up and nothing to install — you ask, and you see which system the agent queries to answer you.
The data is synthetic and the company doesn't exist. What isn't simulated is the mechanics: same tools, same loop and same traceability as in a real implementation. And the same principle in all of them: the agent queries the system of record, never a loose file.
General management, sales and finance
Sales by region and product line, commercial pipeline, P&L, cash flow, receivables, inventory and the Microsoft 365 calendar.
What you'll see: Ask why margin dropped in one region and watch the agent chain two queries before answering.
Finance and controlling
A dashboard with revenue, EBITDA, margin, profit, cash and overdue receivables, and an agent that explains what it shows and draws whatever you ask for.
What you'll see: Ask it to "chart EBITDA against cash flow": nobody programmed that chart — it builds it from the data on the server.
Legal and corporate secretariat
Clauses by topic, two contracts compared line by line, expiry dates and prioritized risks — everything citing contract, version and clause.
What you'll see: Try pasting a contract into the chat: it refuses, and explains why. That's what separates an assistant from a system a lawyer can use.
Procurement and supply
A comparison table by total cost of ownership, each quote's validity, the supplier's delivery track record and RFP drafts with weighted criteria.
What you'll see: Ask for the laptop comparison: the cheapest quote turns out to be the most expensive, and the agent shows you the arithmetic.
Agribusiness: field management, irrigation and crop health
The multispectral mosaic of a 375-hectare farm with its problem zones marked, the evidence and money at risk behind each finding, projected yield, and variable-rate irrigation and fertilisation prescriptions.
What you'll see: Ask it what disease the asparagus has: it refuses to diagnose from an image and explains why. It is the only demo where the data is a picture.
For teams with many documents and little memory: legal, consulting, communications, sales.
The content César Balarezo publishes about AI, turned into a graph that answers what was said, when, and how each piece connects to the rest.
What you'll see: a 3D map that grows over time, topic territories the system found on its own, and each topic's neighborhood on click.
Your area isn't here? They all came out of the same engine: what changes is which systems the agent points at and what it knows how to do with them. If you want to see it on yours, and we'll build it around your case.
In a real implementation these tools point at your ERP, your CRM, your contract repository or your procurement system. Book 30 minutes and we'll go through your case.