Problem Statement
Sari-sari owners want online income but freeze at “ano sasabihin ko sa live?” Marketing apps are English, need data, and push paid boosts. After floods, restock capital is thin; every bundle must move.
Proposed Solution
Snap shelf: vision tags items, app suggests 3 bundles (pang-baon, pang-ulan, pang-handaan) with prices ending in -9, then writes a 60-second live script (“Mga suki, 3 Ariel + 1 Downy, ₱149 lang!”) plus Shopee/TikTok caption with hashtags. Orders via chat or SMS fallback; utang ledger links in one tap.
Target Market
- Primary users: Sari-sari owners, market vendors, teen helpers running lives.
- Deployers: Coops, youth digital aides.
How AI Is Used
- Shelf tagger: Vision lists sellable items from cluttered shelves.
- Copywriter: LLM writes budol-free Taglish scripts that fit 60 seconds.
PLDT / Smart / ePLDT / Rokid Angle
- Smart: Live bundles over mobile data; SMS orders when data drops.
- ePLDT: Micro-store hosting for coops.
Demo Scenario (3 minutes for judges)
- Snap mock shelf; bundles appear with prices.
- 60-second script plays with timer.
- Shopee caption copies in one tap.
- SMS order arrives on basic phone.
- Close: “Mula estante hanggang live.”
30-Hour Build Plan
| Hours | Task | Role |
|---|---|---|
| 0–3 | Mock shelves + scripts | All |
| 3–10 | Tagger + copywriter | AI lead |
| 3–10 | Bundler + SMS | Backend |
| 10–20 | Seller UI | Frontend |
| 20–26 | Tindera phrasing test | All |
| 26–30 | Pitch, disclosure | PM / Pitch |
Suggested stack: Flutter + vision tagger, LLM copywriter, SMS fallback. Public data / APIs: None; team mock shelves.
Risks & Privacy Notes
- Honesty: No fake discounts or scarcity lies; prices from ledger. Demo shelves fictional; no real customer data into AI tools.
- IP: All scripts team-written patterns.
AI Use Disclosure (draft)
Uses vision tagging plus LLM copywriting. Coding assistants helped with UI. All demo scripts human-checked for honesty.
Rubric Scoring
| Criterion | Weight | Score (1–5) | Why |
|---|---|---|---|
| Innovation & creativity | 25% | 4 | Shelf-to-live in Taglish is catchy |
| Problem–solution fit | 25% | 4 | Post-flood income need |
| AI application & value | 20% | 3 | Tagging + writing |
| Feasibility | 15% | 5 | Very buildable |
| Impact & scalability | 15% | 4 | Thousands of stores |
| Weighted total | 4.00 |