Wooble
Build your proof

Wooble FastTrack · Level 3: Advanced

Fraud detection analysis project

Fraud and anomaly investigation, at Potli PaySimulation

The rules engine is down and fraud losses jumped. Run the risk policy over a week of payments, UTC timestamps and all.

  • You're the AI-enabled Data & Finance Analyst
  • About 90 minutes, no timer
  • 7 days to hand in
  • Checked the moment you hand in
  • Free
Join the track

Free. Level 3 opens once you pass 2 Level 2 problems. Start at Level 1.

The situation

Rahul MenonHead of Risk, Potli Pay

Hi, glad you're here.

Our fraud losses jumped last week, and our rules engine is down while it's rebuilt, so nothing was flagged automatically. I need you to run our risk policy (it's in the handbook) over last week's transactions, Monday 28 September to Sunday 4 October, by hand or by code.

I need:

  • Every transaction the policy flags, with the rule that flags it.
  • How many transactions that is, the total amount they add up to, and how many users are involved.

Read the policy carefully: each rule is precise on purpose, and plenty of honest customers come close to one.

Use any tools you like, AI included. Then write me a short note: what should we block today, and which rule would you change?

Thanks, Rahul

When you start, you get Potli Pay's handbook and your own data: the same story as everyone else's, with different records, so your answers are yours alone.

What you hand in

  1. 1Flagged transactions: How many transactions the policy flags (a count)
  2. 2Amount flagged: The amounts of every flagged transaction, added up, in rupees (in whole rupees)
  3. 3Users involved: How many different users the flagged transactions belong to (a count)
  4. 4Flags: A table with two columns, txn_id and rule, uploaded as a CSV or pasted.
  5. 5Your working: The spreadsheet, SQL or notebook that produces your answers, as a link anyone can open.
  6. 6A note to Rahul: 200 words at most.
  7. 7How you used AI: Your prompts or the chat, if you used it.

Your figures and your flags are checked against your own data the moment you hand in. The pass mark is 70 of 100, and the amount flagged must be right. Hand in as often as you need; every hand-in is counted.

What passing earns

  • A step to your AI-enabled Data & Finance Analyst · Level 3 badge: 1 of the 2 passes it needs.
  • A badge page that anyone can check, which you can add to LinkedIn as a certification.
  • A project on your Wooble profile made from your hand-in, when you choose to add it.
  • 50 points for the pass.
  • Finish every level for the track certificate: Wooble FastTrack Certified: AI-enabled Data & Finance Analyst.

Level 3 opens once you pass 2 Level 2 problems.

More in Level 3