About EcoLafaek
AI-powered waste management system connecting citizens, authorities, and technology to combat waste challenges in Timor-Leste
Why EcoLafaek Exists
EcoLafaek aims to address Timor-Leste's waste management challenges through innovative technology, community engagement, and data-driven decision-making.
Dili produces over 300 tons of waste daily, with more than 100 tons going uncollected. This waste clogs drainage systems, causing flooding and environmental damage.
Our vision is a cleaner Timor-Leste where communities are empowered to report waste issues, authorities have the data they need to prioritize cleanup efforts, and environmental health improves.
Expected First-Year Impact
Measurable outcomes for Timor-Leste
How It Works
Citizen Reporting
Citizens use our Flutter-based mobile app to report waste with photos and location data. The app works in low-connectivity environments.
AI-Powered Analysis
Our system processes waste images using Amazon Nova-Pro to classify waste types, assess severity, and evaluate environmental impact.
Coordination & Action
Data is used to prioritize cleanup efforts, directing resources to critical areas. The system tracks progress from report to resolution.
Public Dashboard
Our web dashboard provides visualizations of waste data, including interactive maps, analytics, and community leaderboards.
Key Features
Interactive Mapping
Visual representation of waste incidents and hotspots across Timor-Leste with detailed filtering and spatial clustering.
AI-Powered Analysis
Amazon Nova-Pro integration for waste classification, severity assessment, and environmental impact evaluation.
Community Engagement
Citizen participation through mobile app reporting and gamified leaderboards to encourage involvement.
Low-Bandwidth Support
Mobile app optimized for Timor-Leste's connectivity challenges, with efficient image compression.
Our Technology Stack
Built with cutting-edge AWS AI services and modern technologies for an autonomous, scalable waste management system
AI & Machine Learning
Amazon Nova-Pro
Multi-Modal LLM
Advanced AI model for image analysis, waste classification, and autonomous reasoning with multi-round tool calling capabilities
Bedrock AgentCore
Autonomous Agent Runtime
Code Interpreter and Browser tools for autonomous SQL queries, chart generation, map creation, and web scraping
Titan Embed
Vector Embeddings
amazon.titan-embed-image-v1 for semantic search and similarity matching with 1024-dimensional vector embeddings
Frontend Applications
Backend Services
Cloud Infrastructure
AWS Lightsail
Cloud Hosting
Ubuntu 22.04 LTS hosting with 1GB RAM, 2 vCPUs, SSL certificates, and systemd service management
Amazon S3
Object Storage
Scalable storage for report images, generated charts, and static assets with lifecycle management
Vercel
Serverless Platform
Global CDN deployment for Next.js applications with edge functions and automatic scaling
Database & Storage
Development Tools
Architecture Highlights
Multi-round tool calling with AgentCore runtime
Semantic similarity with Titan embeddings
Live dashboards with interactive visualizations
JWT auth, rate limiting, and data encryption
Frequently Asked Questions
Join Us in Making a Difference
Download our app, report waste incidents, and become part of the solution for a cleaner Timor-Leste
Meet Our Team
EcoLafaek is a dedicated team of innovators working to create technological solutions for challenges in Timor-Leste.
Ajito Nelson Lucio da Costa
Founder & Lead Developer
Passionate about using technology to create positive environmental impact in Timor-Leste through innovative solutions.