This Python framework enables the creation of self-hosted, low-code AI chatbots with advanced NLU capabilities. It targets developers and users seeking to build conversational AI without extensive coding, offering a DIY platform with a user-friendly interface for bot development and training.

How It Works

The framework leverages a combination of machine learning models for intent recognition and entity extraction, including spaCy for word embeddings and scikit-learn/TensorFlow for ML-based NLU. It supports zero-shot NLU via LLMs and is developing RAG for knowledge base querying. The architecture is built on Python with FastAPI, MongoDB for persistence, and React/NextJS for the admin dashboard.

Quick Start & Requirements

• Install: Docker or pip install ai-chatbot-framework.
• Prerequisites: Python 3.7+, MongoDB. Docker is recommended for easier setup.
• Documentation: https://github.com/alfredfrancis/ai-chatbot-framework/blob/main/docs/

Highlighted Details

• Fully self-hosted and low-code DIY platform.
• Supports multi-turn conversations and API request fulfillment (tool calling).
• Features persistent memory, context management, and advanced NLU.
• Integrations via REST API, with planned support for Messenger, Slack, and WhatsApp.

Maintenance & Community

The project appears to be actively maintained by a single primary contributor, alfredfrancis. Contribution guidelines are available.

Licensing & Compatibility

The project is licensed under the MIT License, permitting commercial use and integration with closed-source applications.

Limitations & Caveats

WhatsApp and Slack integrations are listed as "coming soon." RAG functionality for knowledge base answering is currently "in development."