Inspiration
The inspiration for Nova came from a very simple, recurring observation in rural China: while city dwellers have endless apps for wealth management, rural families—who have collectively saved over 20 trillion RMB—have almost no way to grow their money safely. We saw that for most villagers, their hard-earned savings just sit in basic bank accounts, slowly losing value, because green investment projects have high entry barriers and complex terms they don't understand. We wanted to build a tool that speaks their language—literally. Our goal was to turn these "dormant" savings into active support for local green projects like solar power, making sure that the people living on the land are the ones actually benefiting from its sustainable development.
What it does
Nova is an autonomous, barrier-free AI financial agent designed to bridge the wealth gap by channeling rural micro-savings into high-yield green investments. It transforms dormant rural capital into active assets through a specialized investment system that lowers entry thresholds for projects like solar stations and carbon credit initiatives. By utilizing a natural language interface, it allows villagers to manage wealth through simple voice interaction, effectively bypassing the digital divide.
How we built it
We developed the core system using Python, integrating the DeepSeek V3 API to handle complex natural language processing and financial decision-making. The architecture incorporates a P-AMM (Private-Automated Market Maker) protocol to manage liquidity for small-scale rural transactions. We also built a custom data engine that pulls real-time satellite weather data via OpenWeatherMap and market indices from Yahoo Finance to create a hybrid oracle for objective risk assessment.
Challenges we ran into
One of the primary challenges was addressing the lack of traditional credit data and financial literacy in rural areas. We had to design a model that could quantify Total Factor Assets, such as household labor structure and health status, which are not typically captured by urban financial systems. Additionally, ensuring liquidity stability for high-frequency, small-amount rural transactions required complex algorithmic balancing to prevent the depletion of investment pools.
Accomplishments that we're proud of
We are particularly proud of creating a system that makes sophisticated green finance accessible to non-tech-savvy users through a colloquial voice interface. Successfully bridging the gap between 20 trillion RMB in dormant savings and sustainable rural revitalization projects is a significant milestone. Furthermore, our ability to integrate diverse data streams—from satellite imagery to global carbon markets—into a unified decision engine represents a breakthrough in inclusive finance.
What we learned
Through this project, we learned that technology must be deeply rooted in social context to be effective; the Sustainable Livelihoods Framework taught us that rural wealth is about more than just cash. We gained deep insights into the mechanics of automated market makers and the potential of large language models to democratize financial expertise. Most importantly, we realized that green investment can serve as a powerful tool for both environmental sustainability and social equity.
What's next for DIC04-GreenHowetownInvestment
Moving forward, DIC04-GreenHowetownInvestment aims to transition from a simulated prototype to a live, blockchain-enabled ecosystem. Our roadmap focuses on integrating Dynamic Blockchain-Enabled Carbon Monetization, allowing rural households to not only earn investment returns from green infrastructure but also tokenize and trade the resulting carbon credits. We plan to expand the system’s data inputs to include high-resolution satellite IoT sensors for real-time asset monitoring and refine our P-AMM algorithms to handle larger-scale, cross-regional liquidity pools. Ultimately, we seek to collaborate with local village governments to pilot the "Gov-Private Hybrid Funding" model, turning Nova into a standardized toolkit for sustainable rural revitalization across different geographical landscapes.
Built With
- carbon-credit-indexing
- deepseek-v3-api
- natural-language-processing
- openweathermap-api
- p-amm-liquidity-protocol
- python
- v0
- yahoo-finance-api
- yfinance
Log in or sign up for Devpost to join the conversation.