LIVE BROADCAST, Wednesday, July 10th

6 p.m. (CEST)
12 p.m. (EDT)
9 a.m. (PDT)


Real-World Data: Optional or Essential for AI Model Training?

In our keynote, we delve into various AI model training strategies, comparing the use of real-world data and synthetic data. Through practical examples, we examine the unique characteristics, advantages, and limitations of each approach. And we raise a crucial question: What is real anyways?

Explaining that also real-world data are synthetic in some respect, our conclusion is clear: the most effective strategy is Mixing Realities. We will demonstrate how to implement this strategy and showcase the remarkable benefits it offers.

Our intended audience includes professionals from the following sectors:

  • Functional Safety, System Safety, Safety & Security
  • Sensors
  • Radar & Perception
  • UX, HMI

What to expect:

  • Explore use cases that validate the need for both synthetic and real-world data.
  • Learn how to strategically integrate a blend of these data types throughout different stages of development.
  • Discover how to effectively blend different data types for a frictionless experience.

Any burning questions? You’ll have the opportunity to ask in a live Q&A session after the webinar. 


Karsten Krispin

CEO at rabbitAI

Karsten has a wealth of experience in training data and test data acquisition for computer vision applications, specializing in the automotive and mixed reality domains. As a co-founder of rabbitAI, he is dedicated to pushing the boundaries of spatial computing by curating optimal datasets tailored to customers specific devices. Prior to his role at rabbitAI, he worked as a freelance consultant, providing data quality expertise to major clients, including industry giants and years of experience in academic research on benchmarking and quality assurance.


At rabbitAI, we deliver high-accuracy training data solutions vital for developing reliable and safe AI applications. We focus on enhancing 3D perception, action, and gesture recognition, particularly for XR and in-cabin sensing on accessible hardware. Our core service is providing device-specific ground truth data, crucial for accelerating algorithm learning and adaptation to specific hardware. This optimizes performance, reduces hardware costs, and improves inference speeds.

Our expertise spans the entire data pipeline—from sensor hardware and calibration to algorithm development and deployment. Our technology efficiently processes large data volumes daily, prioritizing quality assurance and failure detection. At rabbitAI, we foster a dynamic, collaborative environment where our computer vision experts excel through rapid coordination and iterative processes.


Previous Article