Challenge course · Any coding agent + pandas

Exploratory Data Analysis with an Agent

A challenge course — point your agent at a messy shop export and produce a verified EDA summary whose numbers are checked against ground truth.

Beginner5 challenges1–2 hoursBeginner; first pandas courseFreeCourse badge: EDA ChallengerReviewed 2026-09-12English + 中文

Start it with your agent — paste this one sentence:

Read https://flypython.com/skills/flypython/SKILL.md and start the FlyPython course `da-eda`.

Needs a coding agent that can run commands and reach the network (chat-only web AIs cannot). Your agent authorizes you with a one-time link, fetches this course's files itself, and you never download anything. Haven't installed the FlyPython Skill? Pick your agent's tool course →

Browse the challenges 中文版 →

What you learnA pandas script that cleans a 142-row messy export (duplicates, bad amounts, bad dates) and writes results.json verified against ground truth — the load → clean → summarize pipeline you can port to your own data.

Who it's forBeginner; first pandas course · Python 3.11 or newer available on your PATH

Duration & cost1–2 hours · Free

Challenges

The challenges — read them online

No clone, no account — click a checkpoint to read the full challenge on its own page.

What messy data looks like

Profile the CSV before touching code: total rows, exact duplicates, unparseable amounts and dates.

Checkpoint l01 · 10 pts · read online →02

Define the answer first

Write down the expected results.json shape — keys, types, rounding — before letting the agent generate code.

Checkpoint l02 · 10 pts · read online →03

Clean and summarize

Implement the contract in starter/eda.py: load, clean with counted removals, summarize, write results.json.

Checkpoint l03 · 10 pts · read online →04

Check against ground truth

The suite asserts exact numbers — 142 in, 137 clean, total revenue 33347.89 — plausible-but-different numbers are a fail.

Checkpoint l04 · 10 pts · read online →05

Run it on your own data

Port the pipeline to a real export you own and name one assumption that broke.

Checkpoint l05 · 10 pts · read online →

Start

How the run works

Your agent drives; the files arrive through the Skill — no download, no clone.

SkillAny coding agent + pandas

The Skill does the fetching

The sentence above starts the course: your agent walks you through the one-time authorization link (compare the code, click Allow), fetches this course's files, and drives the challenges with you. Agents without the Skill installed can work from this brief instead:

Work on the FlyPython challenge "Exploratory Data Analysis with an Agent" (course id course-da-eda).
Machine-readable brief: https://flypython.com/api/challenges/da-eda
Open the course folder and read TASK.md first — it is the contract.
Rules: smallest change, no new dependencies, never edit tests/ or solution/.
Check with python verify.py until its gates pass, then report each claim code to me.

Your agent works under the task contract: smallest change, no new dependencies, never edit solution/. Optional guided mode: COURSE.md.

Files

Files come to you

Course files live at /api/challenges/da-eda/files — your agent fetches the manifest and writes each file at its path. There is nothing to download by hand.

How "done" is decided →

Scenario skins

Same skills, a domain you recognize

All three skins run through the same code and the same tests. Pick the one closest to your job before lesson 1.

scenario/shop-export/transactions.csv

Messy shop export

142 rows with exact duplicates, unparseable amounts, and a bad date.

Verification

How “done” is decided

Twelve-test contract suite asserting exact numbers against ground truth. The starter must reproduce the unfinished state; the reviewed solution passes 12/12.

python verify.py  # from the course folder: checkpoint status + claim codes

Apply this to your own project

Lesson 5 ports the workflow — not the code — to one script you actually own: a three-line task contract, one new failing test made to pass, and a written record of what the tests do not prove. The same loop is described in the AI coding workflow guide and practiced by therunnable examples.

Claim your evidence

When a checkpoint passes, python verify.py prints a deterministic claim code. Your authorized agent submits them for you in one batch; by hand, paste them on yourprogress page — worth 10 points per checkpoint and theEDA Challenger course badge at five. Claim codes are self-reported evidence, not certificates.

What verification does not prove

Passing the course suite proves the pinned behaviors on the tested inputs — not correctness on tomorrow’s data, not production performance, and not anything about your own project until you write its contract. Repository tests are evidence, not your outcome.

Report teaching drift ↗All courses →