Jameel Ahmed

Healthcare Claims & Fraud Detection Data Platform

Tech Stack: Azure Data Factory, Databricks (PySpark), ADLS Gen2, Azure SQL, Snowflake, Purview

• Designed and implemented a large-scale batch data platform processing 2.5M+ healthcare claims monthly
• Built ADF pipelines to ingest data from 12+ sources (CSV, JSON, EDI, databases) into ADLS (raw zone)
• Developed Medallion architecture (Bronze–Silver–Gold) using Databricks for scalable data processing
• Implemented data quality validation (150+ rules) and error handling with quarantine mechanisms
• Built rule-based fraud detection engine (500+ rules) improving detection accuracy by 42%
• Designed dimensional models and curated datasets for reporting and analytics
• Integrated Snowflake for analytics and Azure SQL for operational reporting
• Implemented data governance using Azure Purview and secure secret management via Key Vault
• Enabled end-to-end monitoring using Azure Monitor and Log Analytics
• Impact:
• Reduced fraud detection time from 45 days → 48 hours
• Improved data quality to 99.7%
• Reduced infrastructure cost by ~28%

Telecom Customer Churn Prevention (Hybrid Streaming + Batch)

Tech Stack: Kafka, Azure Databricks (Spark Streaming), ADF, Cosmos DB, Snowflake

• Designed Lambda architecture (batch + streaming) for real-time churn detection
• Built Kafka-based streaming ingestion handling high-volume telecom event data
• Developed Spark Structured Streaming pipelines for real-time feature computation (
• Implemented batch pipelines using ADF + Databricks for feature engineering and historical processing
• Created customer 360º data model integrating CRM, billing, and interaction data
• Built real-time serving layer using Cosmos DB for low-latency access
• Developed incremental and batch pipelines feeding Snowflake for analytics
• Implemented CI/CD using Azure DevOps and Bitbucket
• Ensured secure architecture using Key Vault, RBAC, and monitoring tools
• Impact:
• Reduced churn rate from 4.2% → 2.5%
• Achieved real-time detection in ~3 minutes (vs 48 hours earlier)

• Built end-to-end data pipelines using SQL, Python, and Azure services
• Designed normalized and analytical database schemas
• Developed stored procedures and optimized queries for high-volume systems
• Supported enterprise reporting and analytics use cases

• Developed data applications using SQL Server and Oracle
• Built data extraction, transformation, and reporting solutions
• Ensured data security and compliance (HIPAA)
• Gained strong foundation in SDLC and data engineering practices