Insight

Create an AI that knows your data with knowledge bases and agents

Create an AI that knows your data with knowledge bases and agents

Hi, this is the Runyour Team.

There is a frustrating moment when delegating tasks to AI: "It doesn't know our company's information."

Whenever you need internal data—like company policies, product manuals, or past meeting minutes—you have to keep finding and attaching documents or copying and pasting content. It’s also a hassle to explain the same background context every time you start a new conversation. 

By using Runyour's Knowledge Base and Agents together, you can streamline this process. 

Simply upload the necessary work materials to the Knowledge Base and configure an agent with specific roles and instructions to reference those materials when generating answers. 

Create an AI that understands the context of your work and performs tasks according to the roles you define. 

STEP 1. Gather your work materials in the Knowledge Base

First, register the materials you want the AI to reference into the Knowledge Base. 

For example, you can use materials such as: 

  • Company policies and operational manuals
  • Product brochures and technical documentation
  • Customer inquiries and FAQ materials
  • Project meeting minutes and research data

By organizing your work materials in advance, you won't have to repeatedly input the same information every time you ask a question.

Instead of the AI relying only on general knowledge, this step creates a foundation that allows it to reference your organization’s data to find and provide the information you need.

STEP 2. Create an agent with a defined role

Once you have your materials ready, it’s time to decide what tasks to assign. 

For an agent, you can pre-configure its role, response style, and rules to follow through a system prompt. 

Here are some examples of how you can use it: 

  • Product Inquiry Specialist
    Answers customer inquiries by referencing product manuals and FAQs.
  • Internal Policy Reviewer
    Answers questions or provides review comments based on internal regulations and work guidelines.
  • Research Assistant
    Summarizes content and drafts reports based on accumulated research data and meeting minutes.

Once you set a role, you no longer need to repeatedly explain “what perspective to answer from” or “what format to use” every time you start a new conversation. 

Store your data in the Knowledge Base and your workflow in the Agent.

By using both features together, you can build a dedicated agent for your recurring work environment. 

STEP 3. Select the AI model that fits your task

You can also choose the AI model your agent will use. 

Since each model excels at different tasks, try selecting the one best suited for the work your agent will handle. 

For example, choose a model strong in long-form processing if you need to handle lengthy documents and large amounts of information, or a model specialized in Korean if you are working with professional documents written in Korean. 

Experiment with various AI models to find the perfect combination for your agent's specific purpose. 

Build an AI environment tailored to your workflow 

By combining knowledge bases with agents, you can go beyond simple Q&A AI and create an AI-powered workspace that reflects your organization's unique data and processes. 

Update your knowledge base as new materials become available, and refine your agent's settings whenever roles or response criteria evolve. Save time on repetitive explanations and file sharing while leveraging your business context more systematically. 

For individual or team use, get started with Runyour AI. If you want to manage members and usage across your company or organization, get started with Runyour Agent.

👉 Create Your Own Agent with Runyour