Problem Statement
MATATAG rollout added lessons but not prep time; teachers write quizzes past midnight while handling 50-60 kids. Photocopy is ₱1/page; internet is weak. AI quiz tools need uploads teachers don’t trust and English outputs pupils don’t grasp.
Proposed Solution
Snap the lesson: OCR extracts text on-device first, then LLM drafts 10 items (8 MCQ + 2 short) with key, difficulty tags, and a remediation line (“Kung 6/10 pababa, ulitin ang talata 2”). Export as large-print handout or SMS-readable version for zero-paper rooms. Teacher approves every item.
Target Market
- Primary users: Public elementary teachers, ALS facilitators.
- B2G: DepEd school heads sharing banks via QR.
How AI Is Used
- Quiz writer: Generates misconception-based distractors, not trivia.
- Remediation writer: Turns weak items into 5-minute reteach scripts.
PLDT / Smart / ePLDT / Rokid Angle
- Smart: SMS quiz share to parents and study groups.
- ePLDT: Division quiz-bank hosting.
- Rokid: Trainee teachers see demo overlay (stretch).
Demo Scenario (3 minutes for judges)
- Snap mock Araling Panlipunan paragraph; quiz appears in 30 seconds.
- Show answer key with source lines.
- Remediation script appears for low scorers.
- Print preview + SMS version shown.
- Close: “Isang litrato, handa na bukas.”
30-Hour Build Plan
| Hours | Task | Role |
|---|---|---|
| 0–3 | Mock lessons + quiz spec | All |
| 3–12 | OCR + writer prompts | AI lead |
| 3–10 | Bank + export | Backend |
| 10–20 | Teacher UI | Frontend |
| 20–26 | Teacher review | All |
| 26–30 | Pitch, disclosure | PM / Pitch |
Suggested stack: Flutter + OCR, quiz-writer LLM, print-friendly export. Public data / APIs: Public MATATAG guides (reference), team mock lessons.
Risks & Privacy Notes
- Copyright: Lesson photos stay on-device; only generated items export. Demo lessons team-written; no real textbook pages into AI tools.
- Quality: Teacher approves all; key cites source.
AI Use Disclosure (draft)
Uses OCR plus LLM quiz generation. Coding assistants helped with export. All demo items human-reviewed.
Rubric Scoring
| Criterion | Weight | Score (1–5) | Why |
|---|---|---|---|
| Innovation & creativity | 25% | 4 | Tagalog quiz + remediation in 30s stands out |
| Problem–solution fit | 25% | 4 | Nightly prep pain |
| AI application & value | 20% | 4 | Generation needs LLM |
| Feasibility | 15% | 5 | Very buildable |
| Impact & scalability | 15% | 4 | Every teacher |
| Weighted total | 4.20 |