Tirth Shah

Enterprise AWS-based data platform supporting Custody, Reconciliation, Fund Accounting, and Securities Services
reporting across APAC, EMEA, and North America.
• Developed and supported cloud-native data platforms processing 2B+ records monthly and powering 50+
business-critical KPIs/KRIs across global Securities Services operations.
• Built metadata-driven ETL frameworks integrating 30+ financial data sources and 50+ source systems across
custody, reconciliation, collateral management, and fund accounting domains.
• Developed multi-stage AWS data lake architectures (Raw → Trusted → Refined) using S3, Lambda, Glue, Athena,
and PySpark for enterprise-scale reporting and analytics.
• Designed configurable transformation pipelines supporting 100+ transformation workflows, automating data
quality validation, enrichment, and KPI generation.
• Managed 100+ AWS Glue table schemas and enabled self-service analytics through Amazon Athena, improving
accessibility of operational and reporting data for global business stakeholders.
• Supported migration from Terraform-based IaC to Environment as Code (EaC) standards while partnering with
Operations teams to deliver data products for APAC, EMEA, and North America stakeholders.

Databricks-based lending and finance reporting platform supporting Auto Loans & Leases business operations.
• Developed and deployed Databricks Lakehouse pipelines using PySpark, SQL, and YAML-driven transformations to
support Auto Loans finance and reporting workloads.
• Implemented Medallion Architecture-based ingestion and transformation frameworks across raw, curated, and
business consumption layers, improving data quality and downstream reporting reliability.
• Built event-driven ingestion pipelines using AWS S3 Event Notifications and Lambda functions to automate source
file onboarding and reduce operational dependencies.
• Refactored legacy Spark SQL workloads into optimized PySpark frameworks leveraging caching, partition pruning,
and predicate pushdown to improve performance and maintainability while supporting SLA-critical monthly
production cycles.

Cloud modernization initiative for enterprise finance and revenue sharing platforms.
• Migrated 20+ enterprise production pipelines from a legacy Hadoop-based Java/Spark framework to Databricks
Lakehouse, rewriting workloads in PySpark and reducing technical debt by 40%.
• Designed Medallion Architecture-based data platforms with event-driven ingestion using AWS S3 Events, Lambda,
and Apache Airflow, reducing manual intervention by 85% and improving data availability latency by 60%.
• Optimized large-scale PySpark workloads using partition pruning, caching, and broadcast joins, reducing execution
time by 35% and improving cluster resource utilization.
• Supported CI/CD and infrastructure modernization initiatives, contributing to migration from Terraform-based
provisioning to Environment as Code (EaC) standards and reducing configuration drift by 50%.
• Collaborated with business, SRE, and engineering teams to deliver cloud-native finance data solutions while
maintaining 99.9% SLA compliance across critical reporting workflows.
• Coordinated migration planning, release management, and production deployments for Databricks, AWS, Airflow,
Python, and SQL-based finance platforms, accelerating modernization delivery timelines by 30%

Supported Meta's Data Engineering & Privacy organization by developing and maintaining ETL pipelines for privacy
compliance and regulatory reporting initiatives.
• Designed and deployed ETL data models and workflows to production using Apache Airflow and Talend, ensuring
reliability across privacy-critical datasets.
• Translated compliance requirements from the Data Privacy team into scalable, production-ready Python and SQL
pipelines.
• Provided on-call production support and performed root cause analysis for pipeline failures.

• Designed and implemented scalable data ingestion frameworks using StreamSets and Hadoop ecosystems,
optimizing processing strategies for datasets ranging from thousands to millions of records.
• Built end-to-end data ingestion pipelines from heterogeneous sources (GCP, Oracle, Netezza, SQL Server) into
Azure Data Lake Storage, leveraging the appropriate framework based on record volume thresholds.
• Managed multi-frequency data refresh schedules (daily, weekly, monthly, yearly) and performed root cause
analysis on pipeline failures across both StreamSets and Hadoop stack components.

• Developed interactive Power BI dashboards consolidating data from multiple business units, enabling leadership to monitor KPIs in real time and reducing manual reporting effort by automating previously spreadsheet-driven workflows.
• Conducted end-to-end exploratory data analysis (EDA) on large, structured datasets using Python (Pandas, NumPy) and SQL, surfacing actionable business insights and presenting findings directly to senior stakeholders.

• Built and integrated front-end modules for a major aviation client using Angular 4, HTML, CSS, and TypeScript, delivering UI components connected to backend services via RESTful APIs within a cross-functional agile team.