Outdated procedures
The new fee has come into force, but the service script still shows the old one, and each channel answers differently.
Memory for Banks and Fintechs
Products, fees, rules and procedures become a reviewed knowledge base, with an owner and a review deadline. Customer service, back office and AI agents answer customers from the current version.
The problem
Banks and fintechs deal with many rules, internal and from regulators, that change often.
The new fee has come into force, but the service script still shows the old one, and each channel answers differently.
Service agents ask product and operations teams about what is already written in some manual, and the queue of questions grows.
An agent that serves customers handles CPF, card and phone numbers, and nobody knows what was logged or who approved each action.
Before and after
From the published procedure to the answer a customer gets.
| Without the company's memory | With Kluco |
|---|---|
| Without the company's memoryThe new procedure arrives by email and the old one stays in the script | With KlucoThe current procedure is in the knowledge base, with an owner and the date of its last verification |
| Without the company's memoryTwo rules say different things and nobody notices | With KlucoAutomatic review points out contradictions, duplicates and overdue articles |
| Without the company's memoryEach channel answers customers differently | With KlucoAI agents and staff answer from the same reviewed knowledge base |
| Without the company's memoryPersonal data shows up in AI logs | With KlucoEmail, CPF, CNPJ, phone and card numbers can be masked for each agent |
| Without the company's memoryNobody knows who approved an agent's action | With KlucoThe audit log records who approved or rejected each action, with the step-by-step view of the run |
What goes into memory
Documents come in from Google Drive, OneDrive, SharePoint, file upload, API or webhook, and become articles that AI keeps up to date.
Use cases
The most common use cases in banks and fintechs, with the part of the platform that does each one.
Internal rules and procedures live in one knowledge base, with an owner, a review deadline and a version history.
BrainThe agent serves customers on WhatsApp and the other channels with what is in the reviewed knowledge base, and hands over to a person what it does not know.
AgentsThe automatic review points out contradictions, duplicates and expired articles, and the proposals wait in the queue with the before and after.
BrainConfiguration changes and Chat use are recorded, with CSV export, and every agent run keeps its step by step.
LLM Management and ChatWhen the company turns it on, the agent masks personal data in its logs and answers.
AgentsWhen AI writes or updates an article, the change can wait for the approval of whoever owns the topic.
BrainBrain · the memory
The Brain turns products, rules and procedures into articles with an owner, a review deadline and a version history. Every answer shows the passage and its verification date.
Agents · AI agents
Service agents answer on WhatsApp and the other channels with what is in the reviewed knowledge base, and hand over to a person what they do not know. Each one queries only the documents chosen for it.
LLM Management and Chat
Customer service, operations, product and risk use the models their profile allows, in the Chat, which draws on the company's memory.
Control
The company sets up each control below in the console.
FAQ
The current procedure sits in one knowledge base, with an owner and a review deadline. When it changes, everyone queries the new version, and every answer shows the document and its verification date. When the deadline passes, the article shows as overdue.
No. It answers with what is in the reviewed knowledge base and hands over to a person what it does not know. Each article defines whether agents that talk to customers can read it, and sensitive actions can require a person's approval.
No. It organises the rules the company has already set, delivers the current version to whoever asks and records how AI is used. Interpreting the rules and making decisions stay with the company's teams.
With access profiles, data isolated per company, an audit log and the option to mask personal data in the logs and answers of each agent. The Security page covers each control.
Next step
In the demo, we show the procedure knowledge base, automatic review and a service agent with personal data masking.