Engineering Real-time Event-Driven Processing

Please join Uber + Streamlio for an evening of tech talks and networking dedicated to discussing real-world use cases and applications of event-driven processing.

Uber + Streamlio will be demonstrating their respective approaches to leveraging EDP by explaining why it’s important, what we do with it, and how it solves problems, as well as sharing insights for both building and consuming the technology.

More information on speakers:

Amit Nene (Uber)- In this tech talk, Amit Nene will talk about uFlow, a real-time event processing and feature engineering system developed to combat Fraud across Uber's ecosystem. Specifically, he will discuss the overall architecture of this system, its building blocks Apache Samza/Flink, Kafka, Cassandra, Apache Spark, and Uber’s Michelangelo ML engine, and dive deeper into a couple of real-world scenarios demonstrating how real-time processing is crucial in detecting and blocking fraud..

Karthik Ramasamy (Streamlio)-In this tech talk, Karthik Ramasamy will present a couple of real-world use cases based on previous experiences at Twitter and Yahoo, including dynamic ad pricing and product safety that demand low latency, high throughput and ease of operations. He will discuss how these use cases show a consistent pattern that is present in many real-time use cases, and how these requirements drove the need for an end-to-end real-time solution. Karthik will discuss the underlying core technologies (Apache Pulsar, Apache BookKeeper ,and Twitter Heron) developed at Twitter and Yahoo that form the basis for unification.

Agenda:

6:00pm - Doors Open, Food, Drinks, and Networking

6:30pm - Amit Nene (Uber- Fraud Platform)

7:00pm - Karthik Ramasamy (Streamlio)

7:30- Q&A and Networking

Don't forget to RSVP!