To build an AI-powered retail intelligence platform that delivers real-time shopper analytics, automated inventory visibility, and proactive queue management through on-device AI,enabling retailers to reduce stock-outs, improve customer experience, optimize staffing, and increase operational efficiency while maintaining privacy and minimizing cloud dependency.

The receipts 🧾

Every red flag on this statement, in full. These are the four places it bites.

1. Exhibit A

   Retail vision analytics is a saturated commercial category, so a competent build looks like existing products and the innovation bar is high

2. It gets worse

   Shelf-compliance and out-of-stock detection need product-level recognition that is genuinely finicky and often the weak feature

3. Still reading?

   Bundling shopper analytics, stock detection and queue prediction invites shallow breadth across all three

4. And the finisher

   The differentiation rests almost entirely on the on-device privacy angle, so without leaning hard into that it is undistinguished

The damage report 📉

Every score this statement earned, and what each one actually costs you.

• Feasibility

  3/5

  Buildable. Not comfortably. There is a week in here you have not planned for yet.

  Person detection, tracking, dwell analysis and shelf out-of-stock detection are all standard computer vision runnable on edge hardware, and public retail and shelf datasets exist, but doing all three well plus queue prediction on-device is a broad build, and shelf compliance detection in particular needs product-level recognition that is genuinely finicky.

• Innovation scope

  2/5

  Nothing here is new. Your only edge is execution — and execution is also everyone else's only edge.

  Retail vision analytics — footfall, dwell, out-of-stock, queue monitoring — is a mature commercial category with established products, so beyond the on-device and privacy angle there is little genuinely new, and the description bundles the standard feature set.

• Clarity

  4/5

  The ask is unambiguous, which quietly removes your favourite excuse.

  The capabilities are enumerated clearly — shopper analytics, dwell time, out-of-stock detection, shelf compliance, queue prediction — and the on-device and privacy requirements are explicit, so the target is well defined.

• Acceptance potential

  2/5

  The numbers do not like you. Bring something the numbers cannot see.

  It is buildable and the demo is clean, but retail vision analytics is a saturated commercial category with mature products, the innovation is thin beyond the on-device privacy angle, and a judge will see a competent reimplementation of existing retail-analytics tools rather than something distinctive.

• Effort

  Massive

  A semester of work wearing a hackathon costume. Something is getting cut; decide what now, not in week five.

  Detection and tracking, dwell heat mapping, out-of-stock and shelf-compliance recognition, queue prediction and a dashboard, all on-device, is a broad multi-feature build.

• Demo-ability

  Easy

  Easy to demo — and so is everyone else's. Working is the floor here, not the achievement.

  Dwell heat maps, a flagged empty shelf and a queue-congestion prediction on store footage are immediately legible business outputs that a retail judge grasps at once.

• Data

  None supplied

  No dataset comes with this one, so every accuracy figure you quote is a number about labels you invented.

  Nothing is provided with the statement. You are sourcing, cleaning and labelling it yourself, and that work is invisible in the demo but very visible in the questions.

The demo they will have already seen 🪞

Somewhere around 35–85 teams are heading here, and the description is doing the choosing for most of them. They will read the same brief, reach the same architecture, and build a version of the same demo you are planning. Being correct is the floor. If your five minutes could be swapped with the team before you and nobody in the room would notice, you have not picked badly — you have built predictably, which costs exactly the same and hurts more.

What survives 🛡️

The ground worth standing on when the questions start.

• Person detection, tracking and dwell analysis are all standard and runnable on edge hardware with public datasets available
• The on-device, no-video-leaves-the-store design is a genuine privacy strength and the one real differentiator
• Dwell heat maps and out-of-stock alerts are immediately legible business outputs for a retail judge