Prachetan Reddy Soddum

About Prachetan Reddy Soddum

I build the data infrastructure that helps banks catch fraud before it costs customers and the business. As a Senior Data Engineer with 5+ years of experience across banking and financial services, I specialize in engineering scalable, cloud-native data solutions that power fraud detection, risk analytics, and regulatory compliance programs in high-stakes environments. Currently supporting the Risk & Fraud team at Associated Bank, I design and maintain enterprise-grade pipelines on Azure Databricks, Azure Synapse, and Azure Data Factory that process millions of transactions to surface fraud signals in real time. My work directly supports Tier 1–3 deposit fraud operations, account takeover (ATO) detection, and new account fraud (NAF) prevention programs. On the compliance side, I bring hands-on experience with BSA/AML frameworks, SAR/FAR filing workflows, Regulation CC (Expedited Funds Availability Act), and audit-ready data lineage — ensuring every pipeline I build can stand up to regulatory scrutiny. My technical toolkit includes Python, PySpark, SQL, Kafka, Spark Streaming, Hadoop, and Zookeeper — and I\'ve led full on-premises to Azure cloud migrations with zero-disruption transitions for mission-critical fraud systems. I\'ve also built model monitoring programs that continuously evaluate fraud detection model performance and trigger escalation workflows when accuracy degrades. Beyond engineering, I partner closely with fraud strategy, compliance, and leadership teams — translating risk policy into data-driven rule logic and delivering executive dashboards that track fraud rates, charge-offs, net losses, and claims data. What drives me: I believe good data engineering isn\'t just about pipelines — it\'s about protecting people. Whether it\'s stopping an account takeover or flagging a suspicious transaction before it clears, the work I do has real impact on real customers.

Experience

1. Senior Data Engineer

   TEK NINJAS

   Jan 2025 — Present · US

   Senior Data Engineer with extensive experience designing and delivering enterprise-grade data infrastructure for risk management and financial fraud teams within a highly regulated banking environment, supporting deposit fraud and account takeover (ATO) detection programs.• Architected and maintained scalable Azure cloud data pipelines using Azure Synapse Analytics, Azure Data Factory, and Azure Databricks to power real-time and batch fraud detection workflows across high-volume financial transaction datasets.• Led end-to-end data migration from on-premises systems to Azure cloud, ensuring data integrity, governance compliance, and zero-disruption transition for mission-critical fraud and risk reporting pipelines.• Engineered and optimized ETL pipelines using Python, PySpark, and SQL, significantly improving processing speed and reducing cloud compute costs, enabling faster fraud signal generation and more efficient data delivery to downstream consumers.• Built a model monitoring and performance tracking program to evaluate fraud detection model efficacy, generate automated findings, and trigger escalation workflows when model performance degraded below defined thresholds.• Developed executive-facing analytics dashboards to surface key risk metrics — including fraud rates, charge-offs, net losses, customer complaints, and claims data — enabling data-driven decision making for senior leadership and the compliance organization.• Partnered with fraud strategy and compliance teams to implement deposit hold rules and fraud prevention controls, translating risk policy into data-driven rule logic that reduced fraud exposure across account opening and transaction channels.• Applied strong working knowledge of BSA/AML regulations, SAR/FAR reporting, identity theft frameworks, and consumer compliance requirements to ensure all data solutions met audit-readiness standards and supported regulatory obligations.

Education

• Karunya Institute of Technology and Sciences

  Bachelor's Degree, Electronics and computers

  2016 — 2020