Vikas Ravula

About

Data engineer with experience at Bloomberg, UBS, and Bank of America building high-volume financial data platforms and services. Owned an end-to-end pipeline processing ~150–200M records/day (Kafka/Cassandra/S3 → Spark/PySpark → Snowflake) with strong data quality controls and Airflow reliability practices, reporting ~99% reliability and major performance gains. Also built large-scale external API ingestion with compliance-minded rate limiting, schema versioning, and quarantine/validation layers.

Key Strengths

• Owned end-to-end financial data pipeline processing ~150–200M records/day
• Built ingestion from Cassandra/Kafka/S3 with incremental loading
• Implemented data quality framework (null checks, deduplication, reconciliation) for regulatory accuracy
• Designed medallion architecture transformations and modeled data into star schemas
• Airflow orchestration with retries/monitoring/SLA alerts achieving ~99% pipeline reliability and ~50% fewer failures
• Improved query/API performance ~40% via Snowflake optimization, caching, and pagination
• Handled schema changes, duplicates, and late-arriving data using dynamic schemas, idempotent loads, window functions, and watermarking
• Spark optimization (repartitioning, broadcast joins, caching) reducing runtime ~30%
• Built external API ingestion at scale with throttling/batching/exponential backoff and version-aware parsing; used quarantine layers to protect downstreams
• Delivered MVP in ambiguous environment by scoping to core use cases, modular design, and Azure DevOps CI/CD with monitoring/log-based debugging

Experience

Education

Skills

Languages