Skip to content

Tutorial index

This page tells a new user which tutorial to pick, what they will build, and how long it should take to get to a first useful result.

Package tour

Audience: new user, student, collaborator

Best environment: docs + notebook

Time to first result: 5-10 minutes

What you build

  • a mental map of the package
  • a first recommended path
  • a tiny demo run

Entry points

  • docs/package-story.md
  • docs/capability-map.md
  • notebooks/01_start_here.ipynb

Canonical panel benchmark

Audience: methods researcher, reviewer, benchmark author

Best environment: CLI + notebook

Time to first result: 10-15 minutes

What you build

  • canonical spec
  • canonical results JSON
  • canonical report

Entry points

  • docs/quickstart.md
  • docs/case-studies/germany.md
  • notebooks/03_canonical_benchmark.ipynb

Bring your own panel data

Audience: applied researcher

Best environment: notebook first, then script

Time to first result: 15-25 minutes

What you build

  • PanelCase
  • protocol config
  • panel report

Entry points

  • docs/tutorials/custom-panel-workflow.md
  • notebooks/04_custom_panel_data.ipynb
  • examples/custom_panel_data_demo.py

Agent-assisted benchmark workflow

Audience: research engineer

Best environment: agent-enabled IDE

Time to first result: 15-25 minutes

What you build

  • agent spec
  • artifact bundle
  • context plan

Entry points

  • docs/tutorials/agent-workflows.md
  • notebooks/05_agent_workflow.ipynb
  • examples/agent_bundle_demo.py

High-traffic public cases

Audience: maintainer, package evangelist, applied researcher

Best environment: docs + notebook + CLI

Time to first result: 10-20 minutes

What you build

  • event-driven case ideas
  • public data fetch plans
  • attention-oriented demos

Entry points

  • docs/high-traffic-cases.md
  • docs/tutorials/github-stars-impact.md
  • docs/tutorials/crypto-event-impact.md