EcoPin Flutter MVP

Team project for the task AI-Assisted Product Design for Future Entrepreneurs at the Erasmus+ BIP "Enhancing Future Skills through AI, 3D Technologies and Robotics" at University of Ruse, Bulgaria.

Steps taken

We took the following steps in order to generate and optimize the idea behind our team project.

1. Generated a problem statement - "We are solving trash in the nature for concerned local citizens in order to reduce pollution."
2. Generated a user persona (see below)
3. Formed a product description (see below)
4. Generated a list of MVP features through a MoSCoW prioritization
5. Came up with the user flow
6. Filled a Lean Canvas
7. Created a competitor comparison
8. Genereted branding
9. Created this prototype
10. Finally created an investor pitch presentation

Project Description

EcoPin is a gamified mobile/web cleanup app. This MVP uses in-memory demo data and mocked AI so the full product journey can be tested before connecting a backend, maps provider, storage, or real AI vision model.

Slogan: Every Pin Makes Nature Win.

User Persona

Name	Carlos Martinez
Age	25
Occupation	Student
Goals	clean up his city, gain rewards and money
Frustrations	volunteer events are infrequent and lack engagement, there is little feedback showing the real impact of the cleaup activities
Needs	motivation to participate in incentives, recognition with badges and rewards

MoSCoW Prioritization

Must Have	Interactive map
	Camera integration with AI (for before and after analyzis)
	Voucher/Rewards store
	Authentication
Should Have	Badges
	Creating trash points
	Quick onboarding
	Anti-cheat
Could Have	Leaderboard
	Daily Streaks
	Donation to incentives and EcoPin app
	Location-based notifications for clean-up spots
Would Not Have	Direct integration with local sanitation department
	Advanced trash sorting with AI

What is included

• Sign in / profile demo flow
• Trash pin discovery and creation
• Real OpenStreetMap map with tappable cleanup pins
• Pin details and cleanup submission
• Mock photo upload fields
• Points, levels, streaks, badges, leaderboard, challenges
• Reward wallet and mock redemption
• Admin metrics data is still modeled, but the admin panel is hidden from the MVP navigation
• AI service layer with mocked cleanup verification, trash classification, impact analysis, cleanup summaries, user insights, hotspot detection, and mission generation

Run

Flutter is not installed in this environment, so this could not be run locally here. Once Flutter is installed:

flutter pub get
flutter run -d chrome

For mobile:

flutter run

AI architecture

AI functionality is abstracted behind AiService in lib/services/ai/ai_service.dart. The app currently uses MockAiService, but the rest of the application calls the interface methods:

• verifyCleanup
• classifyTrash
• estimateImpact
• generateCleanupSummary
• generateUserInsights
• detectHotspots
• generateMission

This keeps the app ready for OpenAI, Gemini, Claude, local models, or a backend AI service later.