Welcome to Radical Whale!
This quickstart guide will help you set up your first workspace, import data, create an AI agent, and process your first dataset. Let’s get started!Prerequisites
Before you begin, make sure you have:- A Radical Whale account (sign up here)
- A CSV file with data you’d like to process (optional - we’ll provide sample data)
- An OpenAI API key (for AI agent functionality)
Step 1: Create Your Workspace
After logging in, you’ll be guided through workspace creation:- Choose a Workspace Name: Pick a descriptive name for your team’s workspace
- Add Description: Briefly describe what you’ll use this workspace for
- Select Plan: Choose the plan that fits your needs (you can start with the free tier)
- Invite Team Members: Add your colleagues’ email addresses (optional)
Step 2: Set Up Your API Keys
Before creating agents, configure your API keys:- Navigate to Variables in the sidebar
- Click Create Variable
- Set up an OpenAI API key:
- Name:
OPENAI_API_KEY - Key:
OPENAI_API_KEY - Value: Your OpenAI API key (starts with
sk-)
- Name:
Step 3: Import Your First Dataset
Radical Whale works with structured data. Let’s import a sample dataset:- Go to Datasets in the sidebar
- Click Import Dataset
- Choose CSV Upload and upload a file with customer data (name, email, company, status)
- Click Import
Radical Whale supports CSV, Excel, and API imports. For details, see Importing
Data.
Step 4: Configure Dataset Columns
After importing, you’ll see your data organized into columns. Let’s configure them for AI processing:- Click on a column header (e.g., “company_name”)
- Click Edit Column
- Configure the column:
- Data Type: Choose the appropriate type (text, number, etc.)
- Instructions: Tell the AI what to do with this column
- Agent Assignment: We’ll do this in the next step
Step 4: Create Your First Agent
Agents are AI assistants that can read and write data in your datasets:- Click Agents in the sidebar
- Click Create Agent
- Configure your agent:
- Name: “Customer Enrichment Agent”
- Model: Select “GPT-4” (from OpenAI)
- Prompt: “You enrich customer records by researching companies and adding insights”
- Click Create
Step 5: Chat with Your Agent
Test your agent by starting a conversation:- Open your agent’s page
- Click the Chat tab
- Ask: “What customers do we have in the dataset?”
- The agent will use its tools to read your data and respond
Agents remember conversation context and can perform multi-step tasks. See
Chatting with Agents.
Step 6: Assign Agent to Dataset Column
Connect your agent to a dataset column for automatic processing:- Go to your customer dataset
- Click Add Column
- Name it “Enrichment”
- In the Agent dropdown, select your agent
- Add instructions: “Research this company and provide industry insights”
- Click Save
Step 7: Process Your Data
Run your agent on dataset records:- Select rows to process
- Click Process Selected
- Watch as the agent researches each company and adds insights to the Enrichment column
You can process records manually, on a schedule, or via API. See Working with
Records.
Step 8: Review Results
After processing completes:- View enriched data in your dataset
- Click any record to see the detailed view with all enrichment data
- Export results as CSV if needed
Step 9: Create a Page
Document your work with Pages:- Click Pages in the sidebar
- Click Create Page
- Add a title: “Company Research Notes”
- Choose an emoji icon (📊)
- Write in the rich text editor
Use AI Content Generation
Generate content with your agent:- Type
/aiin the editor - Select your agent
- Enter a prompt: “Summarize the company research findings”
- Click Generate
Next Steps
Congratulations! You’ve successfully:- ✅ Created a workspace
- ✅ Set up API keys
- ✅ Imported a dataset
- ✅ Created an AI agent
- ✅ Processed data with AI
- ✅ Created a documentation page
Explore More
Core Concepts
Understand workspaces, datasets, agents, tools, and pages
Managing Tools
Give agents access to web search, code execution, and MCP tools
AI Content Generation
Generate documentation with AI using workspace context
Working with Records
Filter, search, and manage dataset records effectively
Need Help?
- Documentation: Explore our user guides
- Support: Email us at [email protected]
Ready to build something amazing? Start creating in your workspace!
Need Help?
If you run into any issues:- Check our User Guides for detailed tutorials
- Review the API Reference for technical details
- Contact our support team for assistance