Andrea Spina | Voxxed Days

Voxxed Days Milano 2019
on Saturday 13 April

Andrea graduated as Computer Engineer at "Università degli Studi di Modena e Reggio Emilia" (Italy), working on his master thesis and co-authoring "Benchmarking Data Flow Systems for Scalable Machine Learning" science paper at DIMA Group, TU, Berlin. Andrea is currently Head of R&D and product backend Team Lead at Radicalbit, Milan. His main works have been focusing streaming technologies, machine learning and security features: he is the co-author of flink-jpmml project and keeps spreading the voice about how to regulate machine learning end-to-end pipelines and streaming applications in meetups, universities and conferences.

See also https://radicalbit.io/

Blending Event Stream Processing with Machine Learning using the Kafka ecosystem

Conference

Kafka comes today as the essential landmark whenever the main task is to write an events stream application, to build a streaming architecture, to move software toward realtime processing. Machine learning is made of several topics: data pre-processing, training, validation, scoring, deployment and so on; given that some of these aspects easily fit in a streaming scenario many others do not because they have been traditionally grown surrounded by historical data and supported by batch architectures (training). After a brief introduction about "streamophobic" artificial intelligence critical topics, the talk will describe how the latter can be applied in realtime by naturally exploiting the Kafka architecture, detailing the proposed technical solutions about online machine learning training and dynamic model scoring with modern oss ML systems as H2O, justifying this effort by listing the outcoming value, as enabling Kappa Architecture and allowing users to manage end-to-end machine learning applications in a pure streaming context. The talk will conclude by presenting the current production use-cases hosting the targeted solutions.

Sign-in
Make sure to download the Android or iOS mobile schedule.