12-04, 18:30–19:00 (UTC), Data/ Data Science Track
Faustream is an open-source tool I developed that bridges the gap between streaming data and real-time predictive analytics. This talk explores how Faustream leverages Python, Kafka, and Faust to handle high-velocity data streams while applying machine learning models in real-time. We'll dive into its architecture, key features, and applications, demonstrating how it can revolutionize data processing across industries.
The ability to process streaming data and make instant predictions is no longer a luxury today—it's a necessity. However, integrating stream processing with machine learning can be a significant challenge for many organizations and ML engineers. This talk introduces Faustream, a solution designed to tackle this problem head-on.
We'll start by exploring the critical need for real-time predictive analytics in high-stakes industries like manufacturing, retail, healthcare, and finance. Through real-world examples, you'll see how Faustream can be applied to process patient monitoring data for instant health insights or analyze financial transactions for immediate fraud detection or handle real-time sensor data in manufacturing.
The core of the talk will demystify Faustream's architecture, showcasing how it seamlessly combines the power of Python, Kafka, and Faust. You'll gain insights into building scalable, real-time ML pipelines that can handle massive data streams without breaking a sweat.
A live demonstration will bring these concepts to life, illustrating how Faustream can be implemented in various scenarios. Whether you're a data scientist yearning for faster model deployment, an engineer grappling with data stream complexities, or an analyst seeking real-time insights, this talk will equip you with practical knowledge to elevate your data processing game.
By the end of the session, you'll have a clear understanding of how to implement real-time predictive analytics on streaming data, along with the tools to start integrating Faustream into your own projects.
No previous knowledge expected
Joseph Oladokun is a Data Scientist and Machine Learning Engineer with extensive experience across healthcare, finance, and software. Currently pursuing a Master's in Information Systems and Business Analytics at Iowa State University, Joseph has worked with companies like Asana, Helium Health, and RataFX, Autochek Africa, where he developed innovative business solutions using Machine learning and predictive analytics. His open-source project, Faustream, aims to make stream processing and machine learning integration easier. Joseph is passionate open source, and application of data to business problems.