Model the unit economics of a street vendor
A food cart or street stall is a business with margins, sales volume, and working capital, often tracked in the owner’s head. Interview a real street vendor respectfully, reconstruct the unit economics with AI stress-testing your model, and offer one genuinely useful suggestion.
Posted by The AI Internships
The work brief
- 01Pick a vendor you already know or buy from. Be respectful, ask for consent, and explain that this is a learning project.
- 02Learn their numbers through conversation: daily ingredients or stock cost, price per unit, units on a good vs. bad day, location rent, permits or other operating fees, and wastage.
- 03Reconstruct the economics: cost per unit, margin per unit, daily profit range, their break-even day. Ask AI to stress-test the model — what costs did you forget? — then flag every number that’s your estimate vs. their statement.
- 04Offer one useful suggestion grounded in the numbers (a bundle, a peak-hour prep change, a QR display) — and record their honest reaction.
What you’ll produce
Submission standard
Submit the interview notes with quotes, the reconstructed economics, and the suggestion with the vendor’s reaction. Include your AI workflow: your stress-test prompts and the holes AI found (or missed). Add a photo of the real vendor's cart or stall (with consent) that matches the business in your notes, and make sure your economics use the numbers they actually stated.
The conversation: their business, their numbers as stated, direct quotes
The conversation: name the vendor's stall and location, their business and numbers as stated, plus direct quotes and first-hand detail. The stall must match your photo.
Written responseRequiredThe reconstruction: per-unit cost, margin, daily profit range, break-even — stated vs. estimated clearly marked
Written responseRequiredYour numbers-grounded suggestion and the vendor’s honest reaction to it
Written responseRequired
You’ll complete these inside your private workspace.
What you must submit as proof
This brief requires evidence an AI can’t fabricate.
A photo of the real vendor’s cart/stall/shopfront you interviewed — no faces
Photograph the cart or stall, not the vendor. Ask permission before you shoot and blur any face in frame. The business type must match your interview notes.
Image uploadRequired
Submissions without this evidence cannot be submitted.
Protect other people in your proof. Blur faces, names, phone numbers and email addresses before you upload, and refer to anyone you worked with by role or number ("Listener 1", "the stall owner"). Your proof is only ever used to check your work — it is never published, never appears on your certificate, and is never shown in your public portfolio.
How your work is evaluated
The passing benchmark is 70/100.
Real fieldwork
22%The conversation has the texture of a real vendor’s reality — specific items, real quotes.
Sound reconstruction
22%The unit economics are correctly computed with honest stated/estimated flags.
Useful suggestion
11%The suggestion follows from the numbers and respects the vendor’s constraints.
Real vendor, consistent numbers
44%A photo shows the real stall matching the interviewed business, and the reconstructed unit economics use the numbers the vendor stated (stated vs estimated clearly flagged). No photo, or a model whose inputs don't match the interview, = fail.
How we grade your AI usage
Using AI is the point — it’s the skill this certificate proves. You’ll answer three short questions about how you used it: what you asked, what was wrong with its first answer, and what you changed. Specific, honest answers score high. “I pasted the brief and submitted the answer” scores near zero.