Stefan has key roles in product development, development infrastructure, and projects with some of Ontotext’s biggest clients. He was the guiding force behind the creation of the Dynamic Semantic Publishing platform and two of its core components, the Concept Extraction Service, and the Recommendations engine.
In his present role, Stefan manages the Semantic Analytics Solutions squad. His mission is to make sure the team builds exciting, easy to use and run, and semantically powered software. The focus is on creating technologies which empower publishers to generate more relevant, interconnected content and personalized semantic recommendations, which helps them increase user engagement.
Approaches for continuous delivery of ML-based text analytics pipelines in Java
What’s a text analytics pipeline made of? What does it do and how? How to effectively process and index millions of documents through a text analytics pipeline? How to manage your corpus to make the processing smooth and predictable? How to performance tune the JVM for the job? How to make the process repeatable?