Merlin Liquidity
AI at work, end to end
Each product opens in its own tab; MERLIN Studio, MERLIN Flows and AI Unboxed show their pages here first. The sign-in lasts a week on this browser.
Most companies are asked to believe in AI before they can see it work on their own business. MERLIN proposes the other way round: take one real operation, run it with AI agents beside the people who do it today, and judge the result on evidence.
A treasury desk's day is short, has firm rules and clear numbers, and every decision can be checked against the books: what is coming in, what must go out, what to invest and what to bring home. That makes it a fair place to test whether AI helps, and to see where it does not.
The same pattern of AI proposing, code checking and a person deciding applies well beyond treasury. MERLIN shows it working end to end on one desk, so the decision to go further is made on what was seen, not on what was promised.
Most companies are asked to believe in AI before they can see it work on their own business. MERLIN proposes the other way round: take one real operation, run it with AI agents beside the people who do it today, and judge the result on evidence.
Treasury is where it starts. A treasury desk's day is short, has firm rules and clear numbers, and every decision can be checked against the books: what is coming in, what must go out, what to invest and what to bring home. That makes it a fair place to test whether AI helps, and to see where it does not.
The same pattern of AI proposing, code checking and a person deciding applies well beyond treasury. MERLIN shows it working end to end on one desk, so the decision to go further is made on what was seen, not on what was promised.
Two outcomes, and the second stands whatever the first says.
The policy and its checks are yours even if you decide not to use AI for treasury. They can check plans from any treasury system or AI model, or a plan made by hand in a spreadsheet, and are run again whenever the policy, the system or the model changes.
MERLIN itself was built by hand: every agent, check and approval written as code for one desk. Studio is the other way to get there. An operation is drawn as a map of nodes and links, each node running a part from a library: a source that reads the books, a step that prepares the figures, an agent that proposes, a check, a person who approves, an act on a system. Published, the map runs.
A map says what an operation does, independent of the framework that runs it, so a design proven here can be carried to LangGraph, CrewAI or a company's own platform. The canvas is open to all; describing a map, publishing and running take a sign-in.