Skip to content
Documentation

Basic Concepts

Querri is a chat-first data analysis platform where AI helps you explore and understand your data through natural conversation. This guide introduces the core concepts you’ll encounter while using Querri.

Projects

Section titled “Projects”

Projects are containers for your analysis work. Each project represents a complete analytical workflow:

  • Self-contained: Every project has its own data, steps, and results
  • Persistent: All your work is automatically saved
  • Shareable: Projects can be shared with team members or publicly
  • Organized: View and manage all your projects from the Projects page

When you start a new conversation with Querri, you’re creating a new project. The project captures everything that happens during your analysis.

Steps

Section titled “Steps”

Steps are individual operations that make up your analysis. Each step represents a single action:

  • AI-generated: The AI creates steps automatically based on your questions
  • Sequential: Steps execute in order, building on previous results
  • Transparent: You can see exactly what each step does
  • Reusable: Steps can reference data from earlier steps

For example, if you ask “Show me monthly revenue trends for 2024,” the AI might create steps to:

  1. Load your revenue data
  2. Filter to 2024 records
  3. Aggregate by month
  4. Create a line chart

Each step shows its status (running, success, or error) and its results.

QDF (Querri Data Frame)

Section titled “QDF (Querri Data Frame)”

QDF is how Querri stores and displays tabular data. Think of it as a smart spreadsheet:

  • Interactive viewer: Scroll, sort, and explore your data
  • Type-aware: Automatically detects numbers, dates, and text
  • Efficient: Handles large datasets smoothly
  • Portable: Results can be downloaded or copied

When a step produces tabular data, you’ll see it displayed in the QDF viewer. This makes it easy to verify results and understand what’s happening at each stage of your analysis.

Tools

Section titled “Tools”

Tools are the AI’s capabilities for working with your data. Querri has several specialized tools:

duckdb_query

Section titled “duckdb_query”

Executes SQL queries to filter, join, aggregate, and transform data. This is the workhorse for data manipulation.

draw_figure_tool

Section titled “draw_figure_tool”

Creates visualizations like bar charts, line charts, scatter plots, and more. Turns your data into visual insights.

cleaner_tool

Section titled “cleaner_tool”

Handles data cleaning tasks like removing duplicates, handling missing values, and standardizing formats.

forecasting_tool

Section titled “forecasting_tool”

Generates predictions and forecasts based on historical data patterns.

The AI automatically selects and uses the appropriate tools based on your questions. You don’t need to know which tool does what—just ask in natural language.

Library

Section titled “Library”

The Library is your central data repository within Querri:

  • File storage: Upload CSV, Excel, JSON, and other data files
  • Data sources: Connect to external databases and services
  • Reusable: Reference the same data across multiple projects
  • Organized: Keep all your data in one accessible place

When you start a new project, you can pull data from your Library. As you work, you can add more sources to the Library for future use.

How It All Fits Together

Section titled “How It All Fits Together”

Here’s a typical Querri workflow:

  1. Start a project by asking a question in chat
  2. The AI creates steps to answer your question
  3. Steps use tools to query, transform, or visualize your data
  4. Results appear as QDF tables, charts, or text insights
  5. You follow up with more questions, building on previous steps
  6. Share your project when you’re ready to collaborate

Understanding these core concepts will help you get the most out of Querri. The platform handles the technical complexity, so you can focus on asking good questions and understanding your data.

Next Steps

Section titled “Next Steps”