This month, we had insane partnerships with big companies, giving out subscriptions to AI tools completely for free. These are tools that will streamline your business's growth and improve your efficiency.

Just by being a patron this past month, you were given access to six paid subscriptions completely for free. Not only do you access these subscriptions, but you gain the ability for Consulting, networking, collaborating with the community, as well as with myself. You get access to daily AI news, resources, giveaways, and so much more.

If you're interested, check out the patreon link in the description below to gain access to these benefits.

In this section, we will showcase how to create AI agents that can work with tabular data, enabling them to read and process numerical values more effectively than large language models. These AI agents will be capable of processing large amounts of numerical data, such as financial statements, and answering questions about a company's performance, pricing structures, and other related data.

To create this AI agent, we will be using Vector Shift, a platform that allows for the easy creation of AI agents with its drag-and-drop UI, without the need for coding. The process will involve the following steps:

1. Input Node and Output Node: We will start by setting up an input node and an output node to define the flow of the automation.

2. Open AI GPT-4 Model: We will utilize the Open AI GPT-4 model, which is the best-performing large language model, to classify the incoming questions and determine whether they are related to pricing or general information about Ford cars.

3. Condition Statement: We will add a condition statement to route the questions to the appropriate processing node, either the Open AI large language model or the CSV query loader.

4. CSV Query Loader: We will create a CSV query loader node to process the numerical data from the Ford financial statement, allowing the AI agent to provide accurate pricing information.

5. Knowledge Base: We will integrate a knowledge base containing the Ford annual report, enabling the AI agent to answer general questions about the company.

6. Merging Outputs: Finally, we will use a merger node to combine the outputs from the different processing nodes and send the final response to the output node.

By following this process, we will create a versatile AI agent that can handle both numerical data queries and general questions about Ford, providing a seamless user experience for customers interacting with the chatbot.

To create an AI agent that can classify questions and process numerical data, we'll follow these steps:

1. Input Node: Start with an input node to receive the user's questions.

2. Condition Node: Add a condition node to classify the questions into two categories: pricing-related or general information about Ford.

3. Open AI GPT-4 Node: Use an Open AI GPT-4 node to classify the questions. Set a system prompt for the model to identify whether the question is about pricing or general information.

4. CSV Query Loader Node: For pricing-related questions, add a CSV query loader node to process the numerical data from a CSV file containing Ford's pricing information.

5. Knowledge Base Node: For general information questions, add a knowledge base node that contains the Ford annual report. This will allow the agent to provide contextual information about the company.

6. Large Language Model Node: Add another large language model node to improve the output generation from the CSV query loader.

7. Merger Node: Use a merger node to combine the responses from the different branches and send the final output to the output node.

8. Deploy as a Chatbot: Once the pipeline is set up, you can deploy it as a chatbot that can be integrated into your website or other platforms, allowing customers to easily access pricing information and general details about Ford.

By following this approach, you can create a versatile AI agent that can effectively classify questions and process both numerical and contextual data, providing a seamless user experience for your customers.

To merge the different branches and process queries, we'll take the following steps:

1. Merge Branches: We have three main branches in our pipeline - the condition node, the CSV query loader, and the knowledge base query. We'll use a "Merger" node to combine the outputs from these branches and send the final response to the output node.

2. Large Language Model Nodes: We've added multiple large language model nodes to handle different types of queries. The first one classifies the query as either about pricing or general information. The second one processes the CSV query, and the third one handles the knowledge base queries.

3. CSV Query Loader: This node is responsible for processing numerical data from the CSV file. It uses natural language SQL to extract the relevant information based on the user's query.

4. Knowledge Base: We've added a knowledge base that contains the Ford annual report. This allows the agent to answer general questions about the company and its financials.

5. Deploy as Chatbot: Finally, we can deploy this pipeline as a chatbot. Vector Shift makes this easy - we can configure the chatbot's appearance, integration options (e.g., website, WhatsApp, Slack), and other settings to make it ready for use.

The key aspects of this solution are the ability to handle both numerical and contextual queries, the use of multiple large language models to specialize the responses, and the seamless deployment as a chatbot. This allows the end-user to interact with a powerful AI agent that can provide detailed information about Ford's pricing and financials.

In this tutorial, we have showcased how to create an AI agent using Vector Shift that can query and process data, specifically financial data for Ford Motor Company. The key highlights of this automation include:

1. Classifying Questions: The AI agent uses an OpenAI GPT-4 model to classify the incoming questions into two categories - pricing-related questions and general questions about Ford.

2. Numerical Data Processing: For pricing-related questions, the agent utilizes a CSV query loader to extract the relevant pricing information from a pre-loaded CSV file containing Ford's pricing data.

3. General Knowledge Queries: For general questions about Ford, the agent leverages a knowledge base containing the company's financial statements to provide informative responses.

4. Merging Responses: The agent combines the responses from the different processing nodes using a merger node to provide a comprehensive answer to the user.

5. Deployment as a Chatbot: The completed automation can be deployed as a chatbot, allowing users to interact with the AI agent through various channels such as a website, WhatsApp, or Slack.

This example demonstrates the versatility of Vector Shift in creating practical AI-powered solutions that can handle both numerical and contextual data. By automating the process of querying and responding to questions about a company's financial performance, businesses can enhance their customer service and provide valuable insights to their clients.