Instructor Demonstrates Coding Agents For Data Analysis
On March 16, 2026, the author ran a three-hour NICAR 2026 workshop demonstrating coding agents for data analysis using Claude Code and OpenAI Codex. The session used GitHub Codespaces, Python, SQLite, and Datasette, and attendees used budget-restricted Codex API keys that incurred $23 in token charges. Workshop materials include exercises for exploring, cleaning, scraping, and visualizing data, and the handout is published for wider reuse.
Key Points
- 1Demonstrates using Claude Code and OpenAI Codex to explore, clean, and visualize datasets in Python.
- 2Highlights practical workflow with GitHub Codespaces and budget-restricted API keys for reproducible, cost-controlled sessions.
- 3Provides ready-to-use exercises using SQLite, Datasette, and Leaflet to accelerate data-journalist implementations.
Scoring Rationale
Practical, actionable workshop showing agents and reproducible tooling + limited depth and single-source blog detail reduces broader impact.
Sources
Public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
