MNEE PULSE

Inspiration

The inspiration for MNEE PULSE comes from a simple gap in the current Web3 ecosystem: many high-quality content creators provide immense value on social media but receive no direct micro-rewards. At the same time, spending stablecoins for daily needs—like a cup of coffee—remains too complex for average users due to gas fees and fragmented UX. We wanted to build a "living ecosystem" where MNEE isn't just a token in a wallet, but a fluid medium of exchange that flows from digital appreciation to real-world utility.

What it does

MNEE PULSE is a dual-utility ecosystem powered by Agentic AI:

Earn (AI Micro-Tipping): An autonomous AI agent scans social feeds (like Farcaster). It uses LLMs to score content quality. High-quality posts trigger an automatic MNEE reward sent directly to the creator's wallet.

Spend (Retail Utility): Users can spend their earned MNEE at physical merchants through a simulated QRIS payment system. The app handles real-time MNEE-to-fiat conversion, making crypto spending as familiar as scanning a standard bank QR.

Gasless Experience: By utilizing a relayer backend, we remove the friction of gas fees, allowing users to earn and spend without needing to hold ETH for transaction costs.

How we built it

We built a full-stack solution designed for scalability and speed:

Frontend: Built with Flutter to provide a smooth, cross-platform mobile experience with integrated QR scanning and real-time transaction tracking.

Backend: A FastAPI (Python) server deployed on Railway for 24/7 autonomous agent operations.

AI Brain: Integrated OpenAI GPT-4o to perform nuanced content analysis and automated scoring.

Blockchain: Integrated the MNEE stablecoin on the Ethereum Sepolia testnet, using web3dart and custom relayers to facilitate gasless transactions.

Data Sourcing: Leveraged social media APIs to feed content to our AI scouts for real-time evaluation.

Challenges we ran into

One of the biggest challenges was ensuring the AI Agent's scoring logic was both fast and accurate to prevent "gaming" the system. We also had to design a robust backend relayer on Railway that could handle the asynchronous nature of blockchain settlements while maintaining a "less than 1 second" feel in the mobile UI. Simulating the QRIS flow required careful mapping of merchant IDs to wallet addresses to ensure funds always land in the right place.

Accomplishments that we're proud of

Successfully deploying a fully autonomous AI agent that can make financial decisions (sending rewards) without human intervention.

Achieving a "Gasless UX" where a user can download the app and start earning/spending without ever knowing what a "Gas Fee" is.

Bridging the gap between Web3 social protocols and real-world retail behavior.

What we learned

We learned that Agentic AI is the missing piece for stablecoin adoption. When money becomes "programmable" and "intelligent," it can move much more efficiently than traditional banking. We also gained deep insights into building high-performance relayers and the importance of optimistic UI updates for blockchain apps.

What's next for MNEE PULSE

We plan to move beyond simulated QRIS by partnering with real-world payment gateways to settle MNEE directly into merchant bank accounts. We also aim to expand our AI agents to support more platforms and introduce "Session Keys" for even more secure autonomous spending limits.

Built With

• dart
• ethereum-(sepolia)
• fastapi
• flutter
• gemini
• mnee-faucet
• mnee-stablecoin
• openai-gpt-4o
• python
• qris-simulated
• railway
• social-media-apis-(farcaster/x)
• typescript
• web3dart