-
-
airflow automation pipelines dags
-
2 api endpoint predict and update actual label for predcition wrong or right for checking model performance in monitoring dash bord
-
mlflow exprements tracking
-
initial basic label dashbord
Autonomous Fraud Detection Agent
Inspiration
Financial fraud is evolving rapidly, while most fraud detection systems remain static and require manual monitoring and retraining. This gap creates delays in detecting new fraud patterns, leading to financial losses and security risks.
We were inspired to build a self-learning autonomous AI system that not only detects fraud but also thinks, learns, and acts independently. The goal was to design an Agentic AI system capable of continuous monitoring, adaptive learning, and automated decision-making — minimizing human intervention and maximizing response speed.
This project aligns with the vision of Agentic AI, where intelligent agents autonomously manage real-world problems.
What it does
he Autonomous Fraud Detection Agent is a self-learning AI system that monitors financial transactions, detects fraud, and takes automated actions without human intervention.
How we built it
Challenges we ran into
Accomplishments that we're proud of
What we learned
What's next for Autonomous Fraud Detection Agent
Built With
- airflow
- catboost
- cicd
- docker
- fastapi
- mlflow
- mlops
- scikit-learn
