Integrated Qualcomm's Location Data with Gimbal's Ad Impression technology, shifting to Google Cloud and BigQuery using Python, Airflow, and Java Microservices. This resulted in the capability to geo-target 24M mobile customers within milliseconds, empowering marketers.• Mentored and guided Data Scientists, Data Engineers, and Data Analysts on Big Data concepts, fostering a knowledgeable and skilled team.• Expanded ad-hoc requests for computational resources using EMR spot instances, enabling Data Scientists to deploy notebooks and process massive datasets instantly, without relying on DevOps resources.• Optimized Druid Clusters for better performance and introduced HBase for efficient caching of known customer devices, resolving performance bottlenecks and speeding up geo-location searches.• Seamlessly integrated the recently acquired Fysical System's in-memory data framework for the outdoor ads API, facilitating the rapid onboarding of clients like Clear Channel, enhancing data enrichments, and expediting data exports.

Designed a comprehensive 360 view model of the customer's e-commerce journey, implementing hourly Service Level Agreements (SLAs) for crucial Key Performance Indicators (KPIs) such as Lifetime Value (LTV), Churn, Conversion Rates, and attributions like app usage and email response rate.• Orchestrated the development of a single sign-on identity resolution model, eliminating duplicate customer profiles, ensuring a unified source of truth for enhanced product development, and facilitating dynamic product recommendations for customers.• Supervised a team of 3 Data Engineers and 1 Analyst in overhauling Fender's Customer Data Platform, reducing monthly costs by more than 50% through the deployment of Segment.io for web and mobile app events and AWS RedShift for real-time analytics, enabling dynamic marketing campaigns for initiatives like abandoned carts and push notifications.• Standardized, transformed, and resolved data inconsistencies through automated schema matching from diverse data sources ensuring compliance with GDPR and CCPA's right to forget policy, and establishing a robust data layer.• Enhanced the productivity of the customer care team by implementing company-wide Slack bots to address billing, subscription, and account-related inquiries, leading to a higher customer satisfaction rate.

Scaled Scopely's game studio data ingestions from millions to 2 billion events per day using Cloudera Hadoop and Airflow for ETL job orchestrations, facilitating targeted customer engagement and real-time game level monetization.• Addressed data governance, security, and privacy concerns by implementing Cloudera Sentry and Kerberos-based authentication, preventing accidental data deletions, enforcing stringent audit controls, and granting granular access capabilities across the organization.• Significantly reduced erroneous data by 90% through an advanced QA Framework based on Spark Streaming, ensuring downstream ETL jobs initiated only after passing strict QA checks set by each game studio.• Led the migration of Tableau-backed dashboards to Apache Impala, reducing batch refresh times for revenue data to under a minute from the previous 6-8 hours, while achieving a 2x improvement in the performance of live SQL queries.

Unified AEG's 18+ million customer data using a custom Big Data ETL framework in Python, Java, and Avro, enabling seamless ticketing data ingestion in minutes.• Established hybrid infrastructure using Amazon EMR and S3 for automatic scaling of on-sale data, resulting in zero data loss and improved analysis of fraudulent transactions.• Utilized Hadoop's data redundancy and security features to ensure PCI-level compliance for AEG's customer data, with backup and disaster recovery capabilities across all data centers.• Alleviated email overload by creating a monitoring framework that notifies product owners about specific reporting needs based on customized filters.

• Drove the initiative of aggregating business intelligence data with 5-minute granularity such as Facebook likes, page impressions, and check-ins for campaigns enabling the marketing team with rich analytical data.• Improved data loading time by architecting a robust ETL framework with Hive schema dependency checks resulting in clean data sets and lower failure points.• Utilized agile methodologies to align data warehouse projects with system automation projects for rapid deployments of code as well as latest RPM packages across multiple servers.• Tuned Java Heap Size parameters, number of task trackers, and optimized Hive queries doubling data collection capacity reaching over 40 simultaneous application events at the rate of around 1.7GB per day.

• Optimized database performance by tuning indexes, partitions, SQL queries, stored procedures and Linux configurations. Identified and resolved bottlenecks enabling clients to serve content to more than 55M visitors per month. • Saved over $200K in disaster recovery budget by managing implementations of comprehensive backup and high availability database replication solution for over 180 production MySQL instances.• Developed and supported custom MySQL Data Analytics Warehouse, processing ~253K rows per 15-minute interval with over 900 MB of data per hour serving ads to 600+ global partners.

• Increased MySQL’s combined average throughput from hundreds of queries per minute to thousands of queries per minute using read/write split, multiple shards and automatic failovers providing seamless database connectivity to more than 20 million active gamers.• Designed and implemented high availability MySQL clusters using M-M-S topography supporting custom implementation of OpenX serving a zero latency multi region in-game ad platform.