Cursor on a data science project: great at Streamlit, weak at pandas
A data scientist tried Cursor on a real project. Prototyping, code completion and commenting went well, but pandas code kept failing and it deleted a bunch of data files.
Done withCursorSpotted· by Mike Woodward· Oct 21, 2026
Result Code completion and commenting both graded A
“It did outstandingly well at generating a UI in Streamlit. The code was a little old-fashioned and didn't use the latest features, but it got me to a working solution astonishingly fast.”
Spotted: found in the wild. Details are as reported by the source above and have not been independently verified.
What they did
The writer used Cursor on a real data science project and graded it by area. They had it generate a Streamlit UI, framework code such as saving results to a CSV, documentation, and pandas, scraping and data cleansing code.
How it went
Mixed. The Streamlit UI, boilerplate and documentation went well, but technical code needed careful checking, and some had to be written by hand, such as a pandas query for clubs that only appeared as home or away teams. In GitHub it needlessly tied itself in knots and deleted a bunch of data files.
Try it yourself with Cursor
Build a Streamlit app for [project description] using the data in [file]. Use modern Streamlit features, add code to save intermediate results to CSV, and add comments and a markdown file explaining what the code does. Do not delete or overwrite any files. Show me each change before you apply it, and I will review any statistical or dataframe code line by line.
A starting prompt written by Agents in the Wild. Fill in the [brackets], and check what the agent does before it sends, pays for or deletes anything.
A 42k-line site with packages to upgrade: 19 easy, 42 medium and 14 hard, by Cursor's own sort. Cursor cleared the easy and medium ones, then a background agent moved it from Remix v2 to React Router v7.
A MongoDB-like database for a hobby project, written spec-first by Claude, Kiro, Qwen Coder and Cursor. They read zero lines of the code and checked it through tests. The first version took about two weeks, part time.
In SpaceXAI's own guide, engineer Lingxi Li's bots start Cursor cloud agents, check every PR every 30 minutes and merge the low-risk ones. A teammate shipped 2,000+ PRs in a month, they report.
A messy, roughly 900-line Blender script, abandoned. Cursor turned it into a proper repo for all their 3D designs, then built shapes they described in plain language. "These aren't 1-prompt creations", they add.
A huge, convoluted codebase they'd taken over. One prompt in Cursor's Plan mode found the bugs and planned the fix, then the agent made it, where the client's previous team hadn't.
Every visitor picks a World Cup winner, and countries light up by majority vote. When the live dot layer choked on mobile, they rewrote it in WebGL to keep it smooth.