Piyush Mittal

Verve is a market leader in AdTech domain offering programmatic solutions that connect advertisers and publishers to people in real-time. Orchestrated Language Model (LLM)-driven ad personalization initiatives, resulting in a remarkable 15% increase in conversion rates. Led the Adaptive Demand Shaping (ADS) project, optimized model that filtered low-value traffic to optimize query per second (qps) while aligning supply with demand in auction dynamics. This initiative resulted in a 10% increase in revenue. Orchestrated the deployment of machine learning models into production, resulting in a 25% improvement in operational efficiency. Implemented CI/CD pipelines for entire projects, incorporating continuous model retraining, monitoring, and infrastructure as code practices. Led the integration of models into an in-house A/B test framework, playing a key role in parallel experimental design, data preparation, feature engineering, model integration, and statistical analysis. This effort resulted in optimized product performance and facilitated data- driven decision-making

Leading, defining, and implementing end-to-end modern data platforms in support of analytics and AI use cases. Addressing aspects such as data privacy & security, data ingestion & processing, data storage & compute, analytical & operational consumption, data modeling, data virtualization, self-service data preparation & analytics, AI enablement, and API integrations Define and support data transformation and preparation activities independently and in conjunction with data scientists. Advise on, design, and deploy models and solutions built around deep and machine learning technologies. Research, evaluate, and recommend process improvements, including automated systems and enhanced models in support of activities.