CS Dojo

Data Visualization with Claude Code and Python in 10 Minutes

In a brisk ten‑minute demo, the presenter showcases how Claude Code, Anthropic’s multimodal coding assistant, can orchestrate an end‑to‑end data‑analysis workflow for a personal mortgage decision. Starting with a natural‑language query about fixed versus variable rates in Canada, Claude is prompted to locate historical interest, unemployment and inflation data, pull the figures from Trading Economics, and scaffold a Python 3.12 project that uses Plotly for interactive time‑series visualisation. The session walks through creating a research folder, committing markdown notes to a Git repo, opening files in VS Code, and even generating a JavaScript scraper to export the raw tables as CSV. Key technical takeaways include Claude’s ability to chain commands across tools—search, file system, version control, and code execution—without manual hand‑offs. The demo highlights the use of Plotly for rapid charting, the DAFT multimodal data‑processing library for CSV generation, and the convenience of opening results directly in a browser or GitHub Desktop. By fabricating placeholder data before swapping in the real metrics, the presenter demonstrates iterative development and error‑checking, while also illustrating how Claude can write reusable JavaScript snippets for web‑scraping based on URL patterns. Notable moments feature Claude correctly naming the project “mortgage research,” committing changes with appropriate messages, and producing a “misery index” (unemployment + inflation) that is then added to the visualisation. The speaker emphasizes the speed advantage over traditional Excel workflows and cites a recent GitHub Actions failure that Claude diagnosed in seconds. Metrics from the talk show the Claude Code tips repository nearing 200 stars and the DAFT library already at 5,000, underscoring community traction. The demonstration signals a shift toward fully cloud‑native, AI‑driven development pipelines where a single conversational interface can handle research, data acquisition, coding, version control, and deployment. For analysts and developers, this promises faster prototyping, reduced context‑switching, and the ability to embed sophisticated economic modelling directly into everyday decision‑making processes.