Ajay Sivaraman
Data Engineer
Chennai
Summary
Data Engineer at Cognizant with 3 years of experience designing Azure-based data platforms for a large restaurant chain. Builds end-to-end pipelines from ingestion through secure vendor-facing delivery, with a focus on performance, data quality, and scalable system design. Consistent 5/5 performance rating throughout career.
Overview
3
3
years of professional experience
2
2
Certifications
3
3
Languages
Work History
Data Engineer
Cognizant
Chennai
06.2023 - Current
- Architected a data sharing platform serving 25+ external vendors, processing 50M+ daily records from Azure Synapse, Cosmos DB, and on-premises sources across 5 business domains.
- Built a config-driven ingestion framework managing 125 tables with automated dependency resolution and parallel execution, reducing the largest table load time by 83% through partition-aware Delta Lake merges.
- Designed a dynamic row-level security model scoping each vendor to their authorized data, with fully automated onboarding and PAT token lifecycle management. Offloaded reporting to Databricks SQL Warehouses, cutting Power BI refresh time by 70%.
- Built a silver layer with time-based aggregations, embedded DQ checks, and post-load reconciliation. Mentored 2 junior engineers and served as the primary liaison for 25+ external vendor teams.
- Established near real-time fraud detection pipeline, preventing approximately $40k monthly in restaurant fraud through construction of Databricks pipeline that identifies suspicious refund patterns and notifies helpdesk within 2 minutes before card transactions settle.
- Delivered over 10,000 vendor-scoped files daily in under 10 minutes by developing a distributed PySpark delivery system with partition-level parallelism processing 3,000 incoming files per run. Ensured accurate file routing through reuse of platform RLS model. Designed fault-tolerant tracking for individual files to surface failures post-run without interrupting successful transfers.
- Built production pipelines across different business domains ingesting from Cosmos DB, Microsoft Fabric, and on-premises sources into Databricks and Synapse, including complex SCD Type 2 models and secure PII handling via Azure Key Vault.
- Implemented temporal reconciliation logic for late and early-arriving records across pipelines, keeping downstream data consistent without manual intervention.
- Internal Metadata and Governance Platform : Built end-to-end data lineage across Synapse, ADF, and Power BI using Microsoft Graph and REST APIs, and automated service principal governance across 50+ Azure resources with proactive Logic Apps alerting.
Education
BE - Production Engineering
PSG College of Technology
Coimbatore05-2022
Skills
Databricks
Apache Spark
- Azure Data Factory (ADF)
- Azure Logic Apps
- Rest API
- SQL
Azure Storage
Azure Synapse
Websites
Certification
DP-900: Azure Data Fundamentals
TECHNIQUES
- SCD Type 2 (nested)
- Row-Level Security (RLS)
- Partitioning & Data Skipping
- AES Encryption / Key Vault
- Data Modeling (Kimball)
- ETL / ELT Design
- REST APIs
Timeline
Data Engineer
Cognizant
06.2023 - Current
BE - Production Engineering
PSG College of Technology