Methodologies, techniques, and technologies for the professional and continuous engineering of AI/ML-enabled software systems are full of pitfalls. Initially, the transfer and operationalization of data science notebooks may already pose major challenges, both technically and organizationally, before the evolution and complexity of such systems need to be continuously controlled. Here, we use "MLOps" as an umbrella term that describes all activities and artifacts for this purpose. MLOps promises to facilitate the management of end-to-end processes and the development and maintenance of AI-enabled software systems in a testable, traceable, and reproducible manner.

The aim of the Hot Spot is to exchange experiences on the development of AI-enabled software systems and on the topic of MLOps, and to share concrete (mis)successes from practice regarding the use of related approaches, methods, and tools.

Topics include, but are not limited to:

• Elicitation, specification, and quality assurance of requirements and seamless transition into design (e.g., measurable business metrics)
• Workflows (e.g., data engineering, model management, automated pipelines)
• Frameworks, technologies, and tools (e.g., metadata management of artifacts)
• Quality and governance (e.g., reproducibility, traceability, compliance)

Participation in the workshop is free of charge. The event will be hybrid with invite-only guests on-site and unlimited guests online via YouTube streaming.

Registration

You can book the fortiss Hot Spot Workshop "Experiences in engineering AI-enabled systems" directly at the following link.

Click here to register