Subsurface LIVE Sessions

At MIQ Digital India Pvt. Ltd. we collect and process high-volume TV viewing data and apply machine learning models to help TV networks get the maximum value out of their ad slots.

We use Apache Spark MLlib to model and PySpark for data wrangling and feature engineering with a Kafka-based event-driven microservices architecture. It uses a well-defined data engineering ecosystem of a lakehouse architecture built on top of Delta Engine.

This talk will cover scaling MiQ's TV product to market across >50 advertisers, details of pipeline optimization for data at TB scale, and cost optimizations for model generations and prediction.