Docs
Quick Start

Quick Start

Last updated:

Welcome! In this quickstart, you’ll see how Camber transforms data analysis from hours of manual work into a 5-minute conversation with AI.

Your Demo Environment

You’re automatically connected to a demo Snowflake cluster with real-world sample datasets ready to explore. No setup required—just start analyzing!

Available Sample Databases:

  • SNOWFLAKE_SAMPLE_DATA - Standard business benchmark data (150K customers, 1.5M orders, 6M line items) for e-commerce analytics & performance testing
  • CITIBIKE_DB - Real NYC bike-sharing data (57M+ trips, 2K+ stations, weather data) for IoT analytics & geospatial insights
  • WORLDWIDE_ADDRESS_DATA - Global address dataset (564M+ addresses worldwide with GPS coordinates) for massive-scale geospatial analytics
  • ALESCO_CONSUMER_DATABASE_SAMPLE - Consumer/demographic data for customer segmentation & marketing analytics
  • SNOWFLAKE_LEARNING_DB - Training/tutorial database for Snowflake education & best practices

These datasets let you test Camber’s full capabilities without connecting your own data warehouse. When you’re ready, you can connect your own Snowflake instance.

From Data to Insights

Imagine you’re staring at millions of rows in your Snowflake warehouse—say, years of Citibike trip data. You need answers, but writing SQL queries and Python scripts feels slow. Here’s a better way: connect your warehouse to Camber, ask Nova AI questions in plain English, watch as it analyzes your data on cloud compute, discover insights you can instantly share, and turn the whole process into a reusable AI agent that keeps your team informed!

Workflow from Snowflake Data through Camber Connector, Nova AI with Compute, to Insights and Share Agent

Step 1: Connect and Configure

Log into Camber and navigate to Nova from the main dashboard.

Nova interface with connection status, compute resources, agent selection, and prompt entry

1. Connect & Select Resources Ensure you’re “Connected” (top-left). Select your compute node size: XXSMALL for testing, SMALL+ for real analysis. Enable GPU if needed. 💡 Nova’s AI agent runs compute-intensive tasks on cloud nodes, not your local machine.

2. Select an Agent Type @ to see available agents. Choose @nova.snowflake—this agent knows how to use the pre-installed Snowflake connector to Camber’s warehouse.

3. Enter Your Prompt Ask your question in plain English:

@nova.snowflake what can you tell me about the citibike trip data
in CITIBIKE_DB? I want to see patterns

Step 2: Review Results

Nova generates code, queries your data, and produces results.

Nova analysis showing multiple outputs and artifact viewer dialog

1. Explore Results As Nova analyzes, clickable artifacts appear: data tables, generated code, and visualizations. Everything is auto-saved to your Stash chats folder.

2. Artifact Viewer Click any artifact to open the side panel with code, tables, charts, and logs. Ask follow-up questions—Nova maintains context without re-running queries.

Step 3: Share with Your Team

Create an AI agent from your completed analysis

Transform your chat into a reusable AI agent:

  1. Click "…" menu → “Create Agent”
  2. Configure: Name (Citibike Analyst), Alias (@<username>.citibike), Description
  3. Share with colleagues who can query it: @<username>.citibike What were peak hours last month?

Your agent replicates your methodology on updated data and answers questions about your findings! 🎯

Next Steps

Explore More:

  • 🔌 Data Connectors - Connect more data sources
  • 🤖 AI Agents - Build specialized assistants
  • 📦 Apps - Browse scientific applications
  • 💼 Jobs - Automate workflows

Need help? Join our Slack community! 👋

What is Camber?