Github Link - https://github.com/SMVINAYKUMAR2341/AI-Loan-Advisor Problem Statement: Despite advances in AI-driven lending, millions of loan applicants are rejected every year by opaque “black-box” systems that provide no explanation, no guidance, and no path to improvement undermining trust, financial inclusion, and regulatory compliance.
Financial institutions need a transparent, explainable, and real-time loan decisioning system that not only predicts eligibility accurately, but also clearly communicates why a decision was made, enabling users to understand, improve, and trust the process while allowing banks to meet modern compliance and governance requirements. Inspiration Driven by the vision to democratize credit access, I engineered this system to dismantle the "black box" of traditional lending. My goal was to leverage high-precision AI to provide applicants with extreme transparency and actionable financial intelligence, turning a standardized process into a personalized experience that empowers rather than intimidates.
What it does This solo-engineered ecosystem features a high-fidelity Customer Identity Hub for real-time risk assessment and a secure Bank Admin Dashboard for lifecycle management. By integrating Explainable AI (XAI), it provides users with instantaneous, transparent credit decisions while offering bank officers a streamlined, institutional-grade workflow for document verification, decisioning, and disbursement.
How I built it I architected the entire platform from scratch using a modern, scalable stack: FastAPI for the core logic, React/TypeScript for the dual-dashboard frontend, and PostgreSQL/NeonDB for robust data persistence. The intelligence engine is powered by XGBoost for predictive accuracy and SHAP for real-time model interpretability, all wrapped in a premium UI featuring custom WebGL shaders.
Challenges I ran into Orchestrating secure 3-factor authentication across independent dashboards while maintaining a unified backend state was a significant architectural hurdle. Additionally, translating multi-dimensional SHAP weights into human-readable narratives required the development of a custom interpretive logic layer to ensure absolute clarity for the end-user.
Accomplishments that I'm proud of I am proud of successfully delivering a production-ready, full-stack fintech solution single-handedly. This includes the implementation of an automated, RBI-compliant PDF reporting system, a secure 3-factor admin login, and a WebGL-enhanced UI that significantly elevates the visual standard of traditional financial applications.
What I learned This project deepened my expertise in Financial Risk Modeling and Enterprise Security Architecture. I learned that true technical excellence in fintech lies at the intersection of extreme precision and user-centric transparency where a user understands not just what the decision is, but the why behind it.
What's next for AI Powered Loan Eligibility Advisor The roadmap includes live integration with global credit bureaus for real-time data fetching, the launch of native mobile applications for on-the-go monitoring, and the implementation of blockchain-based decentralized identity (DID) to further secure customer KYC. Customer Credentials : Mobile Number : 8095006741 Customer ID : LA20253834 Email: vinay@gmail.com Password: Vinay@123 Pin: 234124 Link of Customer Page: https://ai-loan-advisor-three.vercel.app/ Admin Credentials : Admin ID: LAAD202501 Email: Vinaykumarsm2341@gmail.com ( case sensitive use exact) Password: Vinay@123 Pin: 234124 Link of Admin page: https://ai-loan-advisor-uaoz.vercel.app/login
Built With
- ai
- ml
- mysql
- postgresql

Log in or sign up for Devpost to join the conversation.