Datasets

Datasets are the foundation of everything you build in Cotera. They connect your business data to AI capabilities, serving as the workspace where agents analyze information and take actions.

What Are Datasets?

Think of a dataset as a live connection to your business data that's been prepared for AI to work with. Datasets pull information from your data warehouse - customer records, transaction data, support conversations, product catalogs, or any other business information you want to analyze or act upon.

But datasets are more than just data views. They're active workspaces where you deploy agents and tools to process that information automatically.

Datasets as Your AI Foundation

The Bridge Between Data and Intelligence

Datasets serve as the connection layer between your raw business data and AI-powered automation. They take the information sitting in your warehouse and make it accessible for intelligent analysis and action.

Where Agents Live and Work

Every agent you create operates within a dataset. The dataset provides the agent with:

• Context: The specific business data the agent needs to analyze
• Structure: Organized information that the agent can reliably process
• Workspace: A defined environment where the agent's outputs are stored alongside your data

Foundation for Tool Integration

Datasets also serve as the launching point for tool connections. Whether you're sending data to external systems or fetching additional information, datasets provide the structured environment that tools need to operate effectively.

How Datasets Connect to Your Business Data

Direct Warehouse Integration

Datasets connect directly to your existing data infrastructure. This means:

• Real-time access: Agents work with your current business data, not static snapshots
• No data movement: Information stays in your systems while becoming AI-accessible
• Familiar structure: Data appears in formats you already understand and use

Flexible Data Selection

You can create datasets from any data in your warehouse - customer records, sales data, inventory systems, marketing campaigns, support tickets, user behavior, survey responses, or literally any information you store. If it exists in your data warehouse, it can become the foundation for intelligent automation.

The key is selecting data that contains the information your agents need to make decisions or take actions.

The Dataset Workflow

1. Connect Your Data

Point a dataset to specific tables or queries in your data warehouse. This creates the foundation that everything else builds on.

2. Deploy Agents and Tools

Add agents as columns within your dataset. Each agent becomes a new dimension of intelligence applied to your data.

3. View Integrated Results

Agent outputs appear as new columns alongside your original data, creating an enriched view that combines raw information with AI insights.

4. Take Action

Use agent findings to trigger automated workflows, generate alerts, or feed other business systems.

Why the Dataset Foundation Matters

Organized Intelligence

By organizing AI work around datasets, you ensure that agents have the right context and information to make good decisions. Random data leads to random results.

Scalable Operations

Datasets let you apply the same intelligent analysis to any amount of data - whether you're processing 100 records or 100,000.

Integrated Workflows

Since agent outputs live alongside your business data, incorporating AI insights into your existing reporting and analysis becomes seamless.

Controlled Environment

Datasets provide boundaries and structure that keep AI operations focused and reliable, preventing agents from making decisions without proper context.

Getting Started with Datasets

The most effective datasets start with a clear business objective. Rather than connecting all your data at once, focus on specific data that relates to a particular business process or decision you want to improve.

Once you have a dataset connected to relevant business data, you can begin deploying agents to extract insights, automate analysis, and take actions based on what they discover.

Your dataset becomes the command center where business data meets AI capability, creating automated intelligence that scales with your operations.