Inspiration
Financial institutions today face a growing challenge: balancing profitability with environmental, social, and governance responsibility. Traditional credit scoring systems focus narrowly on historical financial metrics, often ignoring sustainability risks, long-term climate exposure, and inclusiveness. At the same time, ESG and SDG data remain disconnected from real financial decision-making.
SustainaScore AI was inspired by the need to bridge this gap—embedding sustainability intelligence directly into credit and risk assessment workflows using AI, while maintaining transparency, explainability, and commercial relevance.
What it does
SustainaScore AI is an AI-driven credit and risk intelligence platform that evaluates a company’s creditworthiness by combining traditional financial indicators with sustainability and UN SDG-aligned metrics.
The platform:
Assesses financial risk (revenue, debt ratio, cash flow)
Evaluates sustainability impact (environmental, social, governance indicators)
Generates a unified SustainaScore
Provides explainable, human-readable insights
Enables actionable decisions such as approval, review, or rejection
Supports inclusive assessment for SMEs and underserved entities
How we built it
We designed SustainaScore AI using a modular, multi-agent architecture to ensure scalability, clarity, and explainability.
Frontend: A React-based dashboard for company profiling, data input, score visualization, analytics, and decision support
Backend: FastAPI server handling REST APIs and agent orchestration
AI Agents:
Financial Risk Agent
SDG Intelligence Agent
Decision Fusion Agent
Explanation Agent
Scoring Logic: Weighted fusion of financial and sustainability signals
Communication: REST APIs with JSON-based data exchange
Design Focus: Clean UI, low-barrier interaction, and transparency
The system is cloud-ready and designed to integrate future components such as blockchain audit logs and external SDG data engines.
Challenges we ran into
Aligning financial risk metrics with sustainability indicators in a meaningful way
Designing explainable AI logic that is both technically sound and easy to understand
Balancing inclusiveness with responsible risk assessment
Creating a system that feels realistic and deployable within a limited build timeline
Accomplishments that we're proud of
Built a fully working AI-agent-based system
Successfully integrated financial and SDG intelligence into a unified score
Delivered transparent and explainable credit decisions
Designed a professional, decision-focused dashboard
Created a solution that is both impactful and commercially relevant
What we learned
We learned that sustainable finance is not just about better data, but about better decision systems. AI agents are especially powerful for handling complex, multi-objective trade-offs such as profitability, sustainability, and fairness. Explainability and user trust are just as important as predictive accuracy in real-world financial applications.
What's next for SustainaScore AI
Portfolio optimization using SustainaScore
Blockchain-based audit trails for compliance and trust
Automated SDG and ESG reporting
Climate stress testing and scenario analysis
Inclusion-focused micro-credit and SME lending extensions
Multilingual and region-specific deployments
Built With
- amazon-web-services
- css
- explainable-ai
- fastapi
- git
- html
- javascript
- json
- multi-agent-ai
- postgresql
- python
- react
- rest-apis
- sqlite
Log in or sign up for Devpost to join the conversation.