Monika Hristova

DraftKings set out to build a real-time analytics capability before the use cases were fully defined. Rather than waiting for perfect requirements, we invested in the streaming foundation first — allowing meaningful use cases to emerge from the platform itself. We assembled an initial solution to deliver second-level visibility into transactions, active users, and revenue. As stakeholders gained live insight, new needs surfaced — including real-time liability tracking and deeper financial monitoring — many not originally envisioned. As adoption expanded, we evolved the architecture to Kafka, Databricks Spark Declarative Pipelines and Real-Time Mode, as well as Sigma to improve scalability, flexibility, and governance. In this session, we will review the architectural evolution, key streaming design decisions, and how we blend real-time, historical, and forecast data into a financial intelligence system. Attendees will leave with practical lessons for designing streaming-first systems when requirements are evolving.

At DraftKings, ensuring secure, fair gaming requires detecting fraud in real time with both speed and precision. In this talk, we’ll share how Databricks powers our fraud detection pipeline, integrating real-time streaming, machine learning and rule-based detection within a PySpark framework. Our system enables rapid model training, real-time inference and seamless feature transformation across historical and live data. We use shadow mode to test models and rules in live environments before deployment. Collaborating with Databricks, we push online feature store performance and enhance real-time PySpark capabilities. We'll cover PySpark-based feature transformations, real-time inference, scaling challenges and our migration from a homegrown system to Databricks. This session is for data engineers and ML practitioners optimizing real-time AI workloads, featuring a deep dive, code snippets and lessons from building and scaling fraud detection.