Vijay Shankar

I’m a Big Data Engineer with 5+ years of experience across fraud detection, insurance, and financial services domains. I build and optimize batch and real-time data pipelines at scale, with hands-on expertise in streaming architectures, CDC-based ingestion, and Medallion Lakehouse design. I take ownership across the full data engineering lifecycle—from raw ingestion and transformation through performance tuning, data quality enforcement, and CI/CD deployment across GCP and Azure cloud environments.

In my current Stripe-focused Real-Time Fraud Detection work at Infosys, I engineered high-throughput enterprise CDC pipelines using Debezium into a Delta Lake Medallion architecture on GCS, reducing fraud audit query latency by 40% with Z-Order clustering. I also delivered a high-availability Lambda architecture on GCP Dataflow consuming production events from Kafka and Pub/Sub, streamlined feature delivery into BigQuery, and operationalized batch AI/ML feature ingestion via Vertex AI Pipelines. I’ve enforced enterprise governance using Google Cloud Dataplex, built CI/CD schema-contract validation to prevent breaking changes reaching production, and provisioned secure multi-environment infrastructure with Terraform, VPC Service Controls, and CMEK via Cloud KMS—while optimizing autoscaling and Spark execution to decrease overnight batch runtimes by 35%.