Creator prompt
The idea behind this presentation
so this is the template you have to work on and this is the text you have to make th eppst based on the format ppt i have provided "Here is the complete content for all six slides, mapped to the template's fixed sections. Every claim is one you can defend. Placeholders for portal fields are marked in brackets.
---
## SLIDE 1 — Title + Team Details
**PITCH DECK**
**SAKHI (सखी)**
*Privacy-first cycle tracking for India*
**Team Details**
- Team Name: [from portal]
- Team ID: [from portal]
- Theme: [from portal]
- Problem Statement ID: [from portal]
- Problem Statement Title: [from portal]
**Visual:** Sakhi ring mark, top-right, small.
---
## SLIDE 2 — Idea Title + Proposed Solution
**Idea Title:** Sakhi — Consent-First Menstrual Health Companion
**Proposed Solution**
- **Problem:** Indian women rely on Flo and Clue, but most period apps monetise user data. Flo paid a 2021 FTC settlement for sharing health data with third parties without consent.
- **Solution:** An on-device cycle tracker. No ads, no data selling, no third-party analytics.
- **Two roles, one app:**
- **Tracker** — full dashboard: cycle ring, predictions, logging, insights.
- **Supporter** — read-only summary of only 5 fields the tracker chooses to share.
- **Partner mode, consent-scoped:** QR or 6-digit code pairing. Supporter never sees logs, symptoms, moods, or notes. Revocable in one tap.
- **Prediction engine:** Weighted recent-cycle model, not a flat average. Flags low-confidence predictions instead of guessing.
- **Evidence-graded insights:** FIGO-aligned irregularity flags. Never diagnostic. Every health claim labelled *well established*, *commonly reported*, or *still being studied*.
- **Education layer:** Phase-aware hormone explainers in plain language.
**Novelty:**
- Consent-scoped partner sharing with a type-level 5-field whitelist
- Evidence-graded health claims — no diagnosis language anywhere
- India-first design in a category dominated by Western imports
- Open-source, free forever, no monetisation of user data
**Visual:** One screenshot of the Home dashboard or the partner pairing screen.
---
## SLIDE 3 — Technologies + Methodology
**Technologies to be used**
| Layer | Technology |
|---|---|
| Framework | Flutter (single codebase, Android + iOS) |
| State management | Riverpod 3.0 with code generation |
| Local database | Drift on SQLite |
| Navigation | GoRouter |
| Design | Phase-aware custom theming (Fraunces + Inter) |
| Planned integrations | Health Connect (Android), HealthKit (iOS) — read-only |
| Testing | 174 automated tests, golden-image UI snapshots |
**Methodology / Process Flow**
```
Onboarding (6 steps)
↓
Cycle calibration (length + duration)
↓
Last period anchor
↓
Symptoms + conditions
↓
Permissions (optional)
↓
Prediction engine → Home dashboard
↓
Daily logging loop → Insights
↓
Optional: Partner pairing (QR)
```
**Visual:** Simple flow diagram. Boxes and arrows. No text inside boxes beyond 3–4 words each.
---
## SLIDE 4 — Feasibility and Viability
**Feasibility**
- Core tracking, prediction, partner pairing, and insights are built.
- 174 automated tests passing; analyzer clean.
- Runs on mid-range Android hardware. No server dependency for core tracking.
- Open-source release planned — no ongoing infrastructure cost to the user.
**Potential challenges and risks**
| Risk | Mitigation |
|---|---|
| Prediction accuracy on irregular cycles | Weighted model; explicit low-confidence labels; no false precision |
| Partner data leakage | Type-level 5-field whitelist; logs/symptoms/notes never shareable |
| Regulatory (DPDPA 2023) | Local-first design; no third-party analytics SDKs |
| India market awareness | Open-source distribution; word-of-mouth; no ad spend required |
| Physical device testing of partner mode | Planned; currently validated via automated transport tests |
**Strategy for overcoming**
- Ship the honest version first. Privacy claims must match what the code actually does, at every stage.
- Partner with NGOs and clinics for distribution instead of paid acquisition.
- Open-source the engine so the privacy claim is auditable, not just promised.
---
## SLIDE 5 — Potential Impact on Target Audience
**Target audience:** Indian women and people who menstruate, aged 18–45. Secondary: partners, mothers, and supporters who want to understand without surveillance.
**Social impact**
- Gives users a health tool that doesn't treat their body as a data source.
- Normalises consent-first sharing with partners.
- Reduces the "I bleed every month but don't know why" gap.
**Economic impact**
- Free forever. No subscription wall. No ads.
- Open source — NGOs, clinics, and state health programs can deploy it at zero licensing cost.
- Removes the ₹500–₹1,500/year subscription burden that Flo and Clue charge.
**Health impact**
- Evidence-graded education closes the awareness gap.
- Non-diagnostic irregularity flags prompt doctor visits without fear-mongering.
- Supporter mode improves partner support during difficult cycle phases.
**Visual:** Three-column layout — Social / Economic / Health. One icon per column.
---
## SLIDE 6 — Details / Links of Reference and Research Work
**Clinical and regulatory references**
- FIGO classification of abnormal uterine bleeding (2018) — basis for irregularity flag logic
- WHO guidelines on menstrual health (2022)
- India DPDPA 2023 — data protection framework referenced for architecture decisions
**Industry research**
- Flo Health FTC settlement, 2021 — case study in health-data monetisation risk
- "Menstrual apps and privacy" — IoT-Tests privacy analysis of 17 cycle-tracking apps
**Open-source comparables**
- NerdCycle — iOS-only partner-mode implementation
- OpenGym — self-hosted + hosted-tier business model reference
- Element / Immature / Standard Notes — privacy-first open-source patterns
**Project documentation**
- `docs/research-pcos-flags.md`
- `docs/research-partner-mode.md`
- `docs/research-hormone-phases.md`
- Public repo: `github.com/abhyyy559/Sakhi` (when published)
**Visual:** Six logos or favicons in a grid, with source name below each. Cleaner than a bullet list.
---
## What to do before submitting
1. **Fill the portal fields.** Team ID, Theme, Problem Statement ID and Title. None of these can be guessed.
2. **Pick one screenshot for Slide 2.** If the app runs, screenshot it. If it doesn't, use a mockup from the brand sheet. No fake UI.
3. **Write your disclaimer for beta privacy.** Do not write "everything stays on your phone." Write "no ads, no data selling, no third-party analytics." Those are true today and will remain true.
4. **Shorten every line if a bullet wraps to two lines.** The template says no paragraphs. Two-line bullets are the maximum.
5. **Export to PDF only.** No exceptions per the template rules.
---
One thing to decide before you send me final content: **which problem statement are you anchoring to from the portal?** The framing of Slides 2 and 5 changes depending on whether it's *Women's Safety*, *HealthTech*, *Social Impact*, or *Open Innovation*. Tell me which one, and I'll tune the language to match.
" fix the 6 slides only okay