Skip to main content
This tutorial walks through one full pass of the Qualia loop: point the app at your data, ask a question in plain language, and end up with a notebook, a task graph, and a writeup whose every claim traces back to evidence. Each step below builds on the previous one, so you can follow it end to end in a single sitting.
1

Download Qualia

Get the desktop app from quadrillion.io/download. See Getting Started for the per-platform install details and first-launch prompts.
2

Open your workspace

Launch Qualia and open the folder where your data lives, or connect to a remote workspace if the data sits on another machine. Any CSV, Parquet, Excel, or other data file in that folder is immediately available to the agent — there is no import or upload step.
3

Start with a prompt

Describe what you want in the chat box, for example “explore sales_q3.csv”. If the file is not in your workspace yet, attach it with the paperclip, and if you would rather not start from a blank page, click one of the example prompts below the chat box.
Empty Qualia workspace with the chat box and example prompt cards
4

Watch the analysis happen in a notebook

The agent loads your data and runs the analysis in a real Jupyter notebook, narrating what it found in the chat panel as it goes. Every cell is normal, editable code: rerun it, change it, or add your own — and you can also just describe the change you want in chat and let the agent make it.
Agent writing and running notebook cells with a summary in the chat panel
5

Follow parallel work in the task map

For bigger questions the agent splits the work across parallel agents, each owning its own tasks. Open the task map with the waypoints button in the Agents section of the left sidebar to watch them run as a live graph, where boxes are tasks and claims and the lines between them are dependencies.
You can control the parallelism with the settings icon in the chat box, which sets how readily Qualia splits suitable work across agents — see Research style.
Task map showing tasks and claims as a live dependency graph
6

Run a skill

Skills are saved workflows you can invoke on demand instead of re-explaining a process every time. Press / in the chat box to see the selectable list, and start with one of Qualia’s built-ins such as /dataset-exploration or /optimize.
Skill picker opened with the slash key in the chat box
You can also write your own skills, or import the ones you already have from Claude Code or Cursor. Browse everything you have — skills, rules, MCP servers, and more — with the briefcase button in the top right of the window.
Skills library open in the Agent Customization tab
7

Get a research writeup

Ask in the chat to write up the findings, or click Generate a writeup when the agent offers it. You get a writeup with findings, figures, and numbered citations, which you can edit alongside the notebook and export when it is ready to share.
Generated writeup with figures and numbered citations
8

Trace any claim to its evidence

Click a citation in the writeup to open the knowledge graph, focused on the exact evidence behind that finding. Everything Qualia learns is stored there and persists across sessions, so later work builds on what earlier work established rather than rediscovering it.

​
Next steps

  • Chat & AI — More on prompting and research style settings
  • Notebooks — Cell operations, keyboard shortcuts, and exports
  • Tasks — Deep dive into the task system
  • Knowledge System — Learn about claims and validation
  • Agents — Running parallel experiments
  • Skills — Save, write, and import reusable workflows
  • Writeups — Research documents with citations
  • Autonomous Mode — Extended independent work with Slack updates
  • Rules — Add persistent instructions to guide AI behavior
  • Settings — Configure models, appearance, and more