The setting
Loopcycle runs dockless e-bikes and scooters across Portland, Denver, and Austin. You are a data analyst on the growth team, and the fleet throws off ride-level data that leadership leans on every week.
Every mission on this path happens at the same company, so context carries over the way it does in a real job: the data you cleaned in mission two is the data the finance lead questions in mission four.
The missions
1. data cleaningstarter
Clean the March rides export before Loopcycle's growth review
You are the data analyst at Loopcycle, a dockless e-bike and scooter operator. The growth review is Thursday, and the raw March rides export just landed with the usual mess the fleet systems leave behind.
You deliver: A data cleaning report listing every issue in the rides export, the rows affected, and the rule applied to each.
Scored on: Catches the duplicate, Catches range and value errors, Normalizes city names, Documents each rule.
Working from: rides_march_sample.csv, data_dictionary.md.
2. metric definitionstarter
Pin down the active rider metric before Loopcycle's board deck
Loopcycle's board deck is due Friday, and finance and growth keep reporting different active rider numbers off the same data. Nobody trusts the slide until the metric is nailed down.
You deliver: A metric definition doc specifying each KPI's grain, time window, filters, and inclusion rules.
Scored on: Precise grain and window, Exclusion rules, Resolves the conflict, Handles edge cases.
Working from: metrics_request.md, current_definitions.md.
3. analysis narrativecore
Explain Denver's ridership dip in Loopcycle's weekly numbers
Loopcycle's weekly ridership held steady for months, then Denver started sliding while the other two cities kept climbing. The VP wants to know what happened before she reads about it from the board.
You deliver: A written analysis narrative that identifies the dip, quantifies it, explains the likely cause, and states the caveats.
Scored on: Identifies and sizes the dip, Isolates the cause, Honest caveats, Clear story.
Working from: weekly_rides.csv, price_change_notes.md.
4. BI QAcore
QA Loopcycle's executive dashboard before it ships Monday
The new executive dashboard goes live Monday, and the last one quietly reported wrong numbers for a quarter before anyone noticed. Your job is to catch the defects before leadership sees them.
You deliver: A QA findings report scoring each dashboard tile pass or fail with the specific defect and the fix.
Scored on: Finds the distinct-count bug, Finds the filter bugs, Finds the date boundary bug, Gives concrete fixes.
Working from: dashboard_spec.md, source_extract.csv.
5. executive summarystretch
Write Loopcycle's Q1 executive summary for the leadership review
The Q1 leadership review is the one meeting where Loopcycle decides what to fund next quarter. The CEO reads one page and asks three hard questions. You write the page.
You deliver: A one-page executive summary of Q1 with the headline, the Denver story, and three ranked recommendations.
Scored on: Accurate headline, Correct Denver math, Three ranked recommendations, Executive ready.
Working from: q1_summary.csv, leadership_questions.md.
How the scoring works
Each deliverable is graded against the rubric written for that mission. Separately, every mission on every path is graded on how you used AI, against the same four criteria:
- Understood the task. The learner framed the goal for the assistant clearly instead of pasting the brief and hoping.
- Grounded in the material. The learner directed the assistant into the provided files and based the work on them, not on invented facts.
- Verified the output. The learner checked claims, numbers, or coverage against the source material before submitting.
- Iterated with judgment. The learner refined weak parts of the draft with specific follow-ups rather than accepting the first answer.
Both scores, with the work behind them, go on your proof profile. That is what makes a claim like "I can use AI for data analysis" something an employer can check.