A practical way to use Claude Code for data analysis work

A ’s working pattern does not rely on handing a large CSV file to AI and letting it decide . Early AI use mainly replaced StackOverflow and other help forums, while the person still did most of the coding. was used to build a few personal R Shiny apps.

Claude Code was later used to build pipelines, improve analysis speed and , and handle other supporting work. The workflow has moved as much as possible toward AI-driven steps. The work centers on 3 KPI metrics, but the surrounding context strongly affects what the results mean.

Standard analysis scripts are kept for RCT, , propensity matching, and similar methods; Claude reviews the output, points out interesting findings, discusses them, and then writes a custom script for that specific analysis. Dumping a large CSV into AI can produce a polished report, but the results are hard to trust and hard to reproduce. The available text only confirms that Opus 4.6 had been used up to this point.

Key points

  • AI was used around existing analysis scripts, not as a full replacement for the analysis process.
  • Claude Code helped build pipelines and improve analysis .
  • The workflow depends on 3 KPI metrics, but context around those metrics is needed to interpret results.
  • Standard scripts are used for RCT, , and propensity matching before Claude reviews the output.
  • Large CSV uploads can create polished reports, but trust and are weak.
Read original