Mahesh Ramaswamy Vulli

Team: OpenDMP

• Leading a team of Data Engineers to design and build scalable data pipelines for a Data Management Platform (DMP), enabling syndication to media partners such as The Trade Desk, Facebook, and TikTok
• Architecting ingestion workflows that process user click-stream data from Tapad into the OpenDMP system, deriving attributes, segments, and audiences for downstream activation
• Developed Looker dashboards to monitor pipeline performance and key campaign data points
• Collaborating with SRE teams to instrument Spyglass metrics and implement Grafana-based observability dashboards and alerting
• Driving delivery of customer-centric data products aligned with advertiser requirements

Team: Airflow (WFO)

• Served as Airflow administrator within the Ad-Platforms domain, managing multiple environments and ensuring high availability and operational stability
• Led onboarding of new customers to Apple’s custom Airflow platform, tailoring configurations to meet specific business requirements
• Managed and maintained Apple’s customized Airflow codebase, ensuring alignment with evolving functional and technology needs
• Configured and monitored end-to-end observability using Airflow alert dashboards, proactively identifying and resolving issues
• Provided expert support to customers and internal data engineering teams on Airflow deployments, CI/CD pipelines, repositories, and troubleshooting

Teams: Data Factory – Data Lake & Data Warehouse, Customer 360, Customer Experience

• Led a team of Data Engineers in building Data Warehouses, Data Marts, and customer-facing products on GCP (BigQuery)
• Designed and deployed data pipelines to ingest and transform large-scale datasets into Data Lake and Data Warehouse environments
• Built scalable ETL workflows using GCP services to enable high-performance analytics across diverse data sources
• Established stakeholder notification mechanisms for BigQuery and Airflow DAGs using email and Webex alerts
• Conducted code reviews, mentored engineers, and ensured adherence to best practices for DAGs and BigQuery development
• Implemented Infrastructure-as-Code approach by developing Terraform repositories for GCP resource provisioning
• Integrated CI/CD pipelines with GitHub (Terraform) for automated deployment, incorporating FOSSA and SonarQube scans for security and compliance

• Designed and supervised data pipeline architecture within the Data & Analytics team
• Led code reviews and mentored junior engineers to ensure adherence to development best practices
• Developed end-to-end, scalable data pipelines using Apache Airflow, PySpark, and Beam
• Managed migration of legacy Teradata workflows and objects into BigQuery on GCP
• Implemented data quality checks on ingested datasets to ensure accuracy and reliability

• Developed Python-based credit score models and attributes within an Apache Beam framework
• Led and mentored an offshore team to translate complex business requirements into scalable data models
• Containerized models and deployed them on Google Kubernetes Engine (GKE) for production use
• Built PySpark and Apache Beam jobs to enable distributed processing of large datasets
• Implemented robust data quality checks on ingress data prior to loading into BigQuery
• Provided application support for debugging and troubleshooting deployed models
• Conducted code reviews to ensure maintainability, performance, and adherence to best practices

• Designed, developed, and scheduled data pipelines using Apache Airflow (Cloud Composer)
• Served as code owner, conducting peer reviews and enforcing development best practices
• Ingested data from diverse external sources into Google Cloud Storage and BigQuery
• Developed custom Airflow operators to enable seamless integration with external APIs
• Led migration of legacy Teradata tables and stored procedures into BigQuery
• Built ETL/ELT pipelines using Nexla to streamline and optimize data workflows

• The primary aim of this project is to design and implement Python models for the precise calculation of consumer credit scores.
• A pivotal task within this initiative involves the migration of legacy data flows into modern frameworks.
• As a project lead, I have overseen a team of five developers.
• I am actively involved in the deployment of data processing jobs across various platforms.
• Participating in architecture reviews related to BigQuery Data Warehousing is a key aspect of my role.
• A significant responsibility involves evaluating project requirements and facilitating communication with various stakeholders.
• To support audit activities, I have developed ad hoc scripts capable of executing the models.

• The primary goal of this project is to conceptualize and construct a data warehouse while simultaneously developing robust ETL processes.
• As part of this endeavor, I've engaged in data modeling activities aimed at shaping the structure of the Enterprise Data Warehouse (EDW).
• I've been responsible for the creation and modification of various database objects.
• To guarantee the integrity of our ETL processes, I've meticulously developed ETL testing programs using Python.
• Utilizing both Informatica and Pentaho Data Integration, I've designed and implemented ETL processes.
• My role extends to performing data transformations through a combination of SQL, Python, and ETL tools.
• In pursuit of optimal performance, I've been actively engaged in query optimization and ETL performance tuning activities.

• The primary objective of this project is to advance the backend databases supporting the Ratings applications.
• As part of this initiative, I've been responsible for the execution of maintenance and enhancement change requests.
• I've actively collaborated with Business Analysts to gather project requirements.
• My role has involved extensive work with SQL Server and T-SQL.
• To enable seamless data access, I've created backend APIs using both Python and Java.
• I've designed and implemented shell and Python scripts to automate critical data processing and report generation tasks.
• For efficient scheduling and monitoring of batch jobs, I've utilized Control-M.
• Client: Moody’s
• Project: Ratings applications, Single Ratings Data Source, One Ratings platform