AI WhatsApp Chatbot with Hexabot & Wiki Integration

## Project Title

**Build an AI WhatsApp Chatbot Integrated with a Wiki Knowledge Base using Hexabot AI**

## Project Description

We are looking for an experienced developer to build an AI-powered chatbot on WhatsApp, integrated with a Wiki-based knowledge base. The chatbot should use Retrieval-Augmented Generation, or RAG, to answer user questions based on the content stored in the Wiki.

The project also requires a scheduled workflow that extracts documents and content from Google Drive, processes them, and populates or updates the Wiki automatically. The Wiki will act as the main knowledge base used by the chatbot.

**Experience with Hexabot AI is required.**
Please review the Hexabot repository before applying:
[https://github.com/hexabot-ai/hexabot](https://github.com/hexabot-ai/hexabot)

## Project Objectives

The goal is to build a complete working system where:

1. Content is stored in Google Drive.
2. A scheduled workflow extracts and processes Google Drive content.
3. The processed content is published or synchronized into a Wiki.
4. The Wiki acts as the knowledge base.
5. The WhatsApp chatbot receives user questions.
6. The chatbot retrieves relevant information from the knowledge base using RAG.
7. The chatbot generates accurate, contextual responses through WhatsApp.

## Main Requirements

### 1. Hexabot AI Setup and Customization

The chatbot must be built using **Hexabot AI**. The developer should be comfortable working with the Hexabot architecture, including workflows, channels, plugins, actions, and AI agent/RAG capabilities.

Tasks may include:

* Setting up Hexabot locally or on a server.
* Configuring the WhatsApp channel.
* Creating chatbot workflows.
* Implementing custom actions where needed.
* Connecting the chatbot to the knowledge base.
* Testing the full conversation flow.

### 2. WhatsApp Integration

The chatbot should be available through WhatsApp.

The developer should be able to work with one of the following, depending on the final technical choice:

* WhatsApp Business Cloud API
* Twilio WhatsApp API
* Another reliable WhatsApp provider, if justified

Expected behavior:

* User sends a message on WhatsApp.
* Hexabot receives the message.
* The chatbot identifies the user question.
* The chatbot retrieves the right knowledge from the Wiki/RAG system.
* The chatbot replies through WhatsApp.

### 3. Wiki Knowledge Base Integration

The Wiki will be used as the central knowledge base.

The developer should help select or integrate with a suitable Wiki platform, for example:

* Wiki.js
* DokuWiki
* BookStack
* MediaWiki
* Another API-friendly Wiki solution

The Wiki must be accessible programmatically so that content can be created, updated, and organized automatically.

Expected features:

* Create or update Wiki pages from Google Drive content.
* Organize content by folder, category, or document structure.
* Preserve basic formatting where possible.
* Avoid duplicate content.
* Track updates when Google Drive files change.

### 4. Google Drive Content Extraction

A scheduled workflow must be created to extract content from Google Drive and populate the Wiki.

The workflow should support:

* Connecting to Google Drive API.
* Reading files from selected folders.
* Extracting text from documents such as Google Docs, PDFs, DOCX, TXT, or Markdown, depending on feasibility.
* Cleaning and preparing the content.
* Updating the Wiki automatically.
* Running on a schedule, for example daily, hourly, or manually triggered.
* Logging errors and sync status.

### 5. RAG-Based Chatbot Responses

The chatbot should use RAG to answer questions based on the knowledge base.

The developer should be familiar with:

* Embeddings
* Vector databases
* Semantic search
* Document chunking
* Retrieval pipelines
* Prompting with retrieved context
* Reducing hallucinations
* Handling “I don’t know” cases when the answer is not in the knowledge base

Possible vector database options include:

* Qdrant
* Weaviate
* Pinecone
* Chroma
* PostgreSQL with pgvector
* Another recommended option

The final solution should ensure that the chatbot only answers based on the available knowledge base whenever possible.

## Expected Deliverables

The selected freelancer will be expected to deliver:

* A working Hexabot-based WhatsApp chatbot.
* A configured WhatsApp integration.
* A Wiki-based knowledge base setup.
* A scheduled Google Drive-to-Wiki synchronization workflow.
* A RAG pipeline connected to the Wiki/knowledge base.
* Basic admin documentation explaining how to configure and run the system.
* Deployment instructions.
* Source code for any custom plugins, actions, connectors, or scripts.
* Unit tests where applicable, especially for custom Hexabot actions or synchronization logic.
* A short handover session or technical walkthrough.

## Required Skills

Please apply only if you have experience with most of the following:

* **Hexabot AI** — required
* Node.js / TypeScript
* NestJS or similar backend frameworks
* WhatsApp Business API or Twilio WhatsApp API
* Google Drive API
* Wiki platforms and API integrations
* RAG implementation
* LLM APIs such as OpenAI, Anthropic, Mistral, or similar
* Vector databases
* Docker / Docker Compose
* REST APIs and webhooks
* Background jobs and scheduled workflows
* Good documentation practices

## Nice to Have

The following experience would be a plus:

* Prior experience building AI support bots.
* Experience with multilingual chatbots.
* Experience with self-hosted AI/chatbot platforms.
* Experience developing Hexabot plugins, custom actions, or channel integrations.
* Experience with Wiki.js or similar knowledge management tools.
* Experience deploying production-ready chatbot systems.

## Important Notes

This is not a simple FAQ bot. The project requires a proper architecture where Google Drive, Wiki, RAG, Hexabot, and WhatsApp work together reliably.

The chatbot should be designed to be maintainable and extensible. We want to be able to add more documents, update the Wiki, improve the RAG pipeline, and extend the chatbot workflows over time.

## Application Instructions

When applying, please include:

1. A brief description of your experience with Hexabot AI.
2. Links to previous chatbot, RAG, WhatsApp, or AI automation projects.
3. Your proposed architecture for this project.
4. The Wiki platform you recommend and why.
5. The vector database you recommend and why.
6. Your estimated timeline.
7. Any questions or assumptions you have.

Applications that do not mention experience with Hexabot AI will not be considered.