The potential of vision AI for top-line growth is well documented, but most businesses still fail to extract value from the technology with an 87% project failure rate. This talk shares how data-centric AI can significantly increase the likelihood of success based on what we learned from collaborating with hundreds of successful teams. It will contain practical examples and is suited for beginners as well.

Zielgruppe: AI Strategists, AI Project Managers, AI Product Owners, ML Engineers

Log4j has set the internet on fire and once again showed us how much the entire industry depends on open-source software. For developers, it’s crucial they can trust the open-source applications they use, to be secure. So, let's find a way to help open-source maintainers worldwide to provide the best security measures available. My team and I partnered up with Google’s OSS-Fuzz to open-source new security testing tools that automatically find RCEs and many more vulnerabilities at scale. Here’s how you can benefit and contribute to this project.

Zielgruppe: Developers, DevOps Engineers, Open-Source Maintainers

We have helped hundreds of teams with their vision AI projects — established research institutions like Stanford, global corporations like Bayer, and even one or two FAANG companies. During those conversations, patterns started to emerge. We'll share these patterns in the form of best practices on building vision AI pipelines that work following data-centric principles.

Zielgruppe:AI Strategists, AI Project Managers, AI Product Managers, AI Product Owners, ML Engineers

With their "Data Science to go" Platform paretos provide an end-to-end AIaaS solution for companies to derive insights, predictions and dynamic trade-offs without the need for additional data science resources. In this session we want to have a closer look on the iterative approach, the paretos platform and why it is a competitive advantage in a dynamic and complex world.

Using modern AI approaches we created fully autonomous robot heads that engage in heated social media discussions, completely taking the human out of the loop. The TNG Innovation Hacking Team created a prototype of an end-to-end natural language understanding system, employing techniques such as Speech-to-Text, Conditional Text Generation and Text-To-Speech. In this talk, we will give an introduction to e.g. NLP. We will guide you through the process of scraping social media comments, training a text generation model and building robot heads.

Zielgruppe: Manager, Decision Makers, IT Experts, Software Developers, Software Engineers, Data Scientists