Customer Service Chatbot

Build a customer service chatbot that answers questions about your product and embed it into any website.

Step 1 - Load Data into a Knowledge Table

1. Create a Knowledge Table:

   • Go to >> Project >> Knowledge Table.

   • Create a New Knowledge Table with your desired Table ID (table name) and pick a Text Embedding Model.

2. Upload Documentation:

   • Open the table that you have just created.

   • Upload your files to fill up the Knowledge Table. JamAI Base will process your files into Knowledge Rows which you can quickly search up later.

Step 2 - Create and Configure an LLM Agent

1. Create an LLM Agent:

   • Go to >> Chat Table >> Agents >> New Agent.

   • Add and configure your LLM Agent with the following parameters:

     • Agent ID: Name of your agent.

     • Models: LLM Model.

     • Temperature: Set as needed.

     • Max tokens: Set as needed.

     • Top-p: Set as needed.

     • Customize system prompt (optional): Define the behavior of your LLM Agent.

     • User message (optional): Set a first reply to your LLM Agent.

     • AI response (optional): Conversational opener by your LLM Agent.

Step 3 - Configure the LLM Agent to Use RAG (Retrieval-Augmented Generation)

1. Select the LLM Agent:

   • Click on the LLM Agent to bring up the Table View.

2. Update Configuration:

   • In Table View, click on the more icon of the AI (output column) and open settings.

   • Configure the following settings:

     • k: The number of maximum Knowledge Rows that can be fetched as references during RAG.

     • Reranking Model: Rerank the Knowledge Rows retrieved before passing them into the LLM Agent.

     • Knowledge Table: The table to search for Knowledge Rows as references.

Step 4 - Create a New Conversation

1. Start a New Conversation:

   • Select the LLM Agent that you have created and create a new conversation.

   • You will be brought to the Chat Table interface.

2. Toggle Between Modes:

   • Toggle between Conversation Mode and Table Mode using the toggle bar on the top right of the interface.

3. Start Chatting:

   • Begin interacting with your LLM Agent.

Step 5 - Deploy the Chatbot

1. Run the Chatbot:

   • You can run the chatbot within the Chat Table interface.

2. Embed the Chatbot:

   • To embed the chatbot into your website, you can generate an API call and use the LLM in your website.

Advanced Usage: Enrich Knowledge Rows

1. Add Output Column:

   • After creating a new Knowledge Table, add a new Output Column (>> Action >> Add output column).

2. Setup Knowledge LLM Agent:

   • Configure the LLM Agent to further process the text content to enrich your Knowledge Row using the following template:

     • Column ID: The title of the column.

     • Data Type: str.

     • Models: The LLM models.

     • Temperature: 0.1.

     • Max Tokens: 512.

     • Top-p: 1.0.

     • Customize prompt: Prompt to process the Input columns.

3. Start Using the Knowledge Table:

   • See the LLM agent magically process your Knowledge Row when you upload files to the Knowledge Table.

---

By following these steps, you can create a customer service chatbot using JamAIBase that leverages your documentation to answer user questions effectively.