EcoChain Prime

Digitizing the recycling supply chain via AI and Computer Vision to transform informal waste collectors into bankable MSMEs and provide corporate ESG traceability.

Comment

What's next for EcoChain Prime

Inspiration 💡

We saw a massive disconnect in the waste management supply chain. On one side, informal waste collectors (scavengers and MSMEs/Pengepul) are unbankable and trapped in cash transactions. On the other side, multinational FMCG corporations are desperate for validated, unmanipulated Traceability data to meet strict environmental compliance (EPR/ESG). We realized that bridging this gap required more than just a ledger—it required verifiable trust. Thus, EcoChain was born to digitize this ecosystem using Artificial Intelligence.

What it does ⚙️

EcoChain is an AI-driven B2B2C marketplace and management platform for the circular economy.

  1. Supply Side: It provides smart dashboards for MSME aggregators. Using IoT Computer Vision, we automatically validate waste quality and weight without forcing grassroots scavengers to use smartphones. This digital footprint transforms informal cash operations into legitimate financial records, making these MSMEs eligible for micro-loans (KUR).
  2. Demand Side: Corporations purchase high-margin Traceability Certificates. Because every transaction is validated by AI from the ground up, corporations receive 100% audit-proof ESG compliance data without the massive overhead of manual field audits.

How we built it 🛠️

We built EcoChain using a modern, scalable tech stack:

  • Frontend: React.js with TypeScript, styled using Tailwind CSS and Framer Motion for a sleek, responsive, mobile-first experience.
  • Backend: Node.js & Express with TypeScript. We utilized Drizzle ORM connected to a PostgreSQL database.
  • Authentication & Security: We implemented Better Auth for robust session management and proxy-aware security.
  • AI Integration:
    • Computer Vision (YOLO): Deployed for our Smart Scan to validate materials.
    • Generative AI: Integrated to power our AI Weekly Insight, proactive Auto-Matchmaking algorithm, and interactive chatbot.
  • Infrastructure: Deployed on Railway for auto-scaling PaaS hosting.

Challenges we ran into 🚧

  1. Handling Edge Cases in Proxy Environments: We faced significant challenges with IP rate-limiting loops when deploying our backend behind Railway's proxy. We had to carefully configure trustedProxyHeaders and ipAddressHeaders in Better Auth to parse x-forwarded-for headers and avoid false-positive spam blocks.
  2. Database Migration Conflicts: Rapid iteration led to constraint collisions (e.g., Foreign Key duplication) during CI/CD. We resolved this by writing custom idempotent SQL migrations (using DO $$ ... $$ blocks) to ensure bulletproof deployments.
  3. UI/UX for Non-Technical Users: Designing a dashboard that feels cutting-edge but remains highly intuitive for traditional MSME waste aggregators required heavy iteration on our frontend components.

Accomplishments that we're proud of 🏆

  • Successfully building a fully integrated B2B2C platform that connects grassroots waste collectors with corporate ESG needs.
  • Implementing an AI Auto-Match system that proactively finds the best buyers based on historical data.
  • Overcoming complex deployment hurdles to achieve a stable, auto-scaling production environment on Railway.

What we learned 🧠

We learned that combining hardware concepts (IoT Cameras) with software (LLMs & Fintech) creates a powerful "data moat". We also learned the importance of idempotent database design and proper proxy header configurations in modern cloud deployments. Most importantly, we proved that digital inclusion doesn't mean forcing everyone to use a smartphone—sometimes, passive inclusion through a smart aggregation point is the most effective strategy.

What's next for EcoChain 🚀

Our immediate next steps include finalizing the Green Credit Scoring Engine within our Admin Control Panel to automate loan origination for banks. We also plan to expand our Computer Vision models to recognize

Built With

Share this project:

Updates

Submission history