Virtusa’s primary goal was to develop a robust fraud detection system using machine learning and data engineering on AWS.

To achieve this, Virtusa partnered with AWS, gained a detailed understanding of the telco’s existing systems, and planned the process integration and system migration to modern cloud technologies.

A cross-functional strategic team of data engineers, machine learning experts, and cloud architects were brought together. A series of collaborative planning sessions were held to get a deeper understanding of the project, from data ingestion to real-time fraud detection.
To secure buy-in for the transformation and transition swiftly into the execution phase, key communications were established with all telco stakeholders, vendors, and partner teams.

Execution

The team proposed achieving the outcome by automating and streamlining rule-set management, reducing burdensome operational overhead, and enabling more efficient, cost-effective fraud-prevention measures powered by machine learning. This solution was built using AWS’s cloud-native, event-driven architecture.

Virtusa’s DataOps and DevOps teams processed data in real time to generate alerts for potential fraud using AWS-managed serverless components.

A real-time data ingestion pipeline was designed by Virtusa using native AWS services. To enable the generation of detailed tables and queries for subsequent analysis, network usage data was ingested directly from the SVC provider, followed by automated data processing, metadata generation, and data cleansing. Finally, the formatted data underwent validation, enrichment, and summarization.

The next step was to build and deploy machine learning algorithms for detecting anomalies and outliers in the data. Amazon SageMaker facilitated the running of Jupyter notebooks. To support this, Virtusa created a Jenkins-based DevOps pipeline to provision the necessary AWS infrastructure.

Services

Industries

Insights

Careers

About