Inspiration
The increasing number of cyberattacks and banking fraud inspired us to build SentinelAI. Traditional banking systems often analyze cybersecurity telemetry and financial transactions separately, making it difficult to detect sophisticated attacks. We wanted to develop an AI-powered platform that correlates both data sources to identify suspicious activities, generate dynamic risk scores, and help banks prevent fraud in real time.
What it does
SentinelAI is an AI-driven banking cybersecurity platform that continuously monitors employee logins, cybersecurity telemetry, and financial transactions. It detects insider threats and fraudulent activities, generates AI-based risk scores, explains why an activity is suspicious, and provides real-time alerts through a centralized security dashboard. The platform also offers analytics, attack timelines, and automated incident reporting to support faster decision-making.
How we built it
We developed the frontend using React, Vite, Tailwind CSS, and Recharts to create a modern banking dashboard. The backend was built with FastAPI and Python, while MongoDB Atlas stores transaction and security logs. Our AI models, built with Scikit-learn and XGBoost, analyze user behavior and transaction patterns to detect anomalies. We also integrated Geo Location, IP Reputation, and Gemini APIs to enrich threat analysis and provide explainable AI insights.
Challenges we ran into
One of the biggest challenges was correlating cybersecurity telemetry with banking transaction data to detect complex attacks while minimizing false positives. Designing an intuitive dashboard that presents large volumes of security information clearly and generating explainable AI predictions for security analysts were also significant challenges.
Accomplishments that we're proud of
- Built a centralized AI-powered banking security dashboard.
- Developed real-time fraud detection with dynamic risk scoring.
- Implemented explainable AI to provide clear reasons behind every alert.
- Designed a scalable architecture suitable for banking environments.
- Created a modern and responsive user interface for security analysts.
What we learned
This project strengthened our understanding of cybersecurity, fraud detection, AI-based anomaly detection, full-stack web development, REST APIs, cloud deployment, and data visualization. We also learned the importance of Explainable AI in helping security teams make faster and more informed decisions.
What's next for SentinelAI – AI-Driven Banking Cybersecurity Platform
We plan to integrate SentinelAI with real banking systems and SIEM platforms, support real-time event streaming using Apache Kafka, improve AI models with deep learning techniques, add automated incident response workflows, and expand the solution for FinTech companies, NBFCs, insurance providers, and digital payment platforms.
Log in or sign up for Devpost to join the conversation.