In Peliqan, you can build a custom AI client (harness). Peliqan provides all the components to access data, implement a semantic layer and a permission layer and publish a front-end for your Chatbots.

AI Chatbot capabilities

With Peliqan, you can build AI clients with following capabilities:

• Text to SQL: allow your AI to convert any question in natural language to an SQL query, executed on the Peliqan data warehouse. Your AI can combine data from all your sources (CRM, ERP etc.) and answer complex analytical questions.

• RAG (Retrieval Augmented Generation): allow your AI to search internal knowledge including structured and non-structured data (e.g. Google Drive, Notion etc.) in order to answer any knowledge-related question. More info:

  Implement RAG

• MCP style actions: allow your AI to take actions in your business applications such as scheduling a task, adding draft invoices based on worked hours etc.

• Permission layer: you can implement a role-based permission layer in Peliqan to access data

• Link to external tools: you can allow the AI to provide links to business tools, e.g. opening invoices in the accounting software or opening a query in a BI tool for further analytical processing of a question

Basic example of an AI Chatbot

Below is a basic example of a chatbot, created in Peliqan.

You can pin the chatbot to the Peliqan homepage, or you can Publish it to make it available outside of Peliqan.

• Streamlit script for Peliqan (click to expand)

Example of an AI Chatbot performing Text-to-SQL

Below is an example of a Chatbot that performs Text-to-SQL. The question of the user is converted to an SQL Query which is executed on the Peliqan data warehouse.