Ads Core ML

• Implementing and improving ML models that optimize ad delivery and drive measurable business impact

• Owning the full lifecycle of production ML systems — including system design, data pipeline development, deployment, validation, monitoring, and cost analysis.

• Integrating 1st-party behavioral signals to enhance user conversion prediction models and improve Ads ranking performance.

• Built statistical models to estimate conversions lost due to privacy constraints (e.g., ATT), improving compliance and preserving $100M+ in ad revenue.

• Collaborating cross-functionally with product managers, data scientists, and infra teams to scale ML solutions across Meta's Ads ecosystem.

• Developed Python and GO microservices within Kubernetes to create API endpoints for business and data engineering teams, enabling seamless data integration and processing at scale.

• Engineered and optimized a highly scalable backend API leveraging GO and Redis, resulting in a 90% reduction in API response time and enabling seamless integration across applications throughout the organization.

• Led the design and implementation of end-to-end data pipelines, harmonizing data from various sources into HDFS, ensuring reliable and efficient data processing for analysis; reduced data integration time by 80% and enhanced data consistency across the organization.

• Spearheaded the creation of scalable applications that effectively presented customer experience data

• Engineered and deployed automation tools that significantly streamlined repetitive tasks within the team, boosting productivity and enabling team members to focus on higher-value activities.

• Received two promotions during tenure for outstanding performance and contributions to project success.

• Building end to end data pipelines and performing analysis on data, for various projects within VMware.

• Using technology such as Spark, Hive, and Sqoop to move data from different databases such as MongoDB, PostgreSQL, and HANA and putting into a Hadoop Distributed File System.

• Writing programs to automate the data pipelines described.

• Creating data visualization dashboards with Tableau.

• Technologies used: HDFS, Spark, Kubernetes, Hive, MongoDB, Kafka, Tableau, GO, Python, Sqoop.

• Machine learning research lab focused on unmanned robots/vehicles, and their interactions with humans.

• Implemented target-searching strategies for autonomous robots by integrating Bayesian Loops and Gaussian Mixtures with human feedback

• Created a simulation environment within Unreal Engine 4, using ROS (Robot Operating System), to facilitate the testing and validation of autonomous drone searching algorithms. This innovation provided a controlled and immersive platform for algorithm refinement and optimization