Big Data Developers Meetup “AI and Cloud” (03.04.2019) 18 Uhr

** Mr Pat – from hackathon to reality – Daniel Frick 30 min** ?‍?

Sometimes you have to work smarter not harder – This is how the jury evaluated the app “Mr. Pat”, which was built for the hackathon of the TU Munich 2018. Due to the use of IBM Watson services such as Watson Assistant or Visual Recognition, the app could be built quickly and easily, while other teams tried to build a Visual Recognition all weekend. The success of Watson Services convinced the organizers, so maybe the app will soon become reality? Often the results of hackathons are turned into reality. An example of this would be the OpenSmart app, which was released in a similar form by BMW and will also be presented at the Meetup.

**BIAS / Fairness at AI – Sumit Goyal 30 min** ?‍?

Rapid development of AI and AI based solutions has greatly influenced public and enterprises alike. Nevertheless, there are some key barriers in a widespread adoption of AI. We need measures to be able to trust and explain the decisions from AI, it’s value to the business, and its implications to customers. It is incredibly important to ensure that the AI is free from bias to avoid business reputation risk and legal exposure. Let us have a look how IBM solutions help businesses ensure they are bringing transparency to AI decision-making with unique capabilities such as automatic bias detection, mitigation, and explain ability. These features give insights, both at build and run times, into AI models performance and fairness to better serve users.

**IBM Cloud Databases: Turning Open Source Databases Into Cloud Services – Tim Waizenegger 30 min** ?‍?

IBM Cloud Databases is built as a native Kubernetes application that runs containerized versions of Open Source databases. Databases are provisioned as clusters or master/slave configurations, depending on the database, to offer high availability. In this presentation, we give an overview of the system architecture and show how we integrated the different database systems. We discuss the advantages and disadvantages of running such a service on Kubernetes and provide insight into how we operate this service.

**IBM Academic Initiative and Lab Boeblingen Activities – Martin Mähler (10 min)** ?

IBM has a rich set of tools free of charge for students and faculties. This includes software downloads, cloud services as well as educational offerings (badges) around Cloud, Data Science, Machine Learning or IoT (see also The talk gives a brief overview also on related activities at the Boeblingen Development Lab.