Karini AI Documentation
  • Introduction
  • Getting Started
  • Organization
  • User Management
    • User Invitations
    • Role Management
  • Model Hub
    • Embeddings Models
    • Large Language Models (LLMs)
  • Prompt Management
    • Prompt Templates
    • Create Prompt
    • Test Prompt
      • Test & Compare
      • Prompt Observability
      • Prompt Runs
    • Agentic Prompts
      • Create Agent Prompt
      • Test Agent Prompt
      • Evaluation
    • Prompt Task Types
    • Prompt Versions
  • Datasets
  • Recipes
    • QnA Recipe
      • Data Storage Connectors
      • Vector Stores
      • Create Recipe
      • Run Recipe
      • Test Recipe
      • Evaluate Recipe
      • Export Recipe
      • Recipe Runs
    • Agent Recipe
    • Batch Recipe
    • Databricks Recipe
  • Copilots
  • Observability
  • Dashboard Overview
    • Statistical Overview
    • Cost & Usage Summary
      • Spend by LLM Endpoint
      • Spend by Generative AI Application
    • Model Endpoints & Datasets Distribution
    • Dataset Dashboard
    • Copilot Dashboard
    • Model Endpoints Dashboard
  • Catalog Schemas
    • Connectors
    • Catalog Schema Import and Publication Process
  • Prompt Optimization Experiments
    • Set up and execute experiment
    • Optimization Insights
  • Generative AI Workshop
    • Retrieval Augmented Generation(RAG)
    • Agentic RAG
    • Generative Business Intelligence
    • Intelligent Automation
    • GraphRAG Agent
Powered by GitBook
On this page
  • Build Text-2-SQL agentic assistant
  • Use case:
  • Prerequisites:
  • Step 1: Catalog Creation
  • Step 2: Create an Agent prompt
  • Step 3: Create Agent 2.0 Recipe
  • Step 4: Create a copilot.
  • Step 5: Ask Questions in the chatbot
  • Sample resources
  1. Generative AI Workshop

Generative Business Intelligence

Build Text-2-SQL agentic assistant

This lab focuses on developing an agentic chatbot that converts natural language queries into SQL commands by leveraging catalog schemas and databases. You will integrate NLP models, schema parsing, and automated SQL generation to enable efficient, intelligent, and user-friendly database interactions.

Use case:

  • Enterprise sales advisor chatbot analyzes ticket sales data for various events covering a range of performances, such as musicals, concerts, and other shows.

Prerequisites:

  • Use the Redshift catalog for enterprise sales data.

Step 1: Catalog Creation

  1. Setup Redshift Catalog in Catalog Schema builder.

  2. Click Add new button in the top corner in Catalog Schemas

  3. Enter the name and select Redshift as the connector type.

  4. Enter the credentials.

  5. Save catalog.

  6. Open the saved catalog.

  7. Browse and Import schema.

  8. To generate business metadata, select a table, enable checkboxes, and then click "Generate Metadata." Repeat this process for each table individually.

  9. Publish the catalog.

  10. The catalog is now ready for use in the Prompt Playground.

Step 2: Create an Agent prompt

  1. Navigate to Prompt Playground.

  2. Click the Add new button in the top corner of Prompt Playground.

  3. Select prompt template Enterprise Sales Agent-Redshift.

  4. Increase the Max State Updates from 3 to 20.

  5. Add your natural language question "What is the total sales amount for each seller?" to the Agent Input (mandatory) text box.

  6. Ensure you rename your prompt as loaded from the prompt template before saving it.

  7. Save the prompt. After saving the prompt, the Tool tab becomes visible within the prompt.

  8. Set up a Catalog tool.

    1. Click Add New in the tool section and enter an appropriate name and description.

    2. Select the type Catalog.

    3. Select the catalog created in Step 1.

    4. Save tool.

  9. Set up a Database tool.

    1. Click Add New in the tool section and enter an appropriate name and description.

    2. Choose Database as the type and select Redshift as the connector.

    3. Use Redshift Datawarehouse credentials.

    4. Save tool.

  10. In the Test & Compare tab, choose the preferred models, assess their outputs, and identify the most optimal model. Subsequently, click "Select as Best Answer" to designate it as the Primary Model. Additionally, if necessary, you may assign an alternative model as the Fallback Model. However, selecting a Primary Model is mandatory.

  11. Click "Save prompt run" button provided at the right corner.

  12. Save and publish prompt.

Step 3: Create Agent 2.0 Recipe

  1. Navigate to Recipe.

  2. Click the Add new button in the top corner of Recipe.

  3. Provide a suitable name and description of the recipe.

  4. Select type Agent 2.0.

  5. Setup following tiles.

    1. Chat

    2. Start - Connect the Chat tile to the Start tile.

    3. Agent (Select the Agentic prompt created in Step 2).

      1. In the Agent Tile, select All Messages from the Messages dropdown under the Agent State flag.

    4. Connect the Start tile to the Agent tile.

    5. End - Connect Agent tile to End tile.

  6. Save and Publish the recipe.

  7. Click Test button on Chat tile to test the recipe.

  8. Ask questions in the recipe Test.

    1. What is the average sales amount among top 10 sellers?

    2. What is the total sales amount for top 10 sellers?

    3. What is the total sales amount for each seller?

Step 4: Create a copilot.

  1. Click the "Export" button located at the top-right corner of the recipe interface.

  2. Click Add new button.

  3. Configure the export form by referring to the sample resource recipe provided below.

  4. Save the export form.

  5. The Copilot will be displayed in the Copilot section of the interface.

Step 5: Ask Questions in the chatbot

  1. Go to the Copilot tab and open your chatbot.

  2. Ask questions.

    1. What is the average sales amount among top 10 sellers?

    2. What is the total sales amount for the top 10 sellers?

    3. What is the total sales amount for each seller?

Sample resources

  1. Catalog schema: Enterprise-Sale-Data (Redshift)

  2. Prompt:Enterprise Sales Agent-Redshift

  3. Recipe: Enterprise Sales Advisor

  4. Copilot: Enterprise Sales Advisor

Refer following video to create recipe.

PreviousAgentic RAGNextIntelligent Automation

Last updated 2 days ago