Munich Re: Achieving AI Democratization With Dataiku
Munich Re: Achieving AI Democratization With Dataiku
Dr. Andreas Bayerstadler of Munich Re shares how a close partnership with Dataiku helps build AI democratization through a citizen data scientist program.
Bayerstadler began by talking about the scope and reach of Munich Re. “When it comes to data analytics, we’re serving different lines of business: that includes global life and health insurance, nonlife insurance, and primary insurance. We’re also doing specialty insurance — for example machinery, insurance of private aircrafts — but we mainly work with companies.”
Munich Re’s data analytics team seeks to create value in two ways. First, it supports internal processes like underwriting and claims insurance. Second, it provides services to primary insurers by helping them understand their data, perform risk assessments, and make data-driven decisions. As key elements to create this value, Bayerstadler identified data, people, and technology.
He showed how understanding of both people and technology have led to strong AI democratization.
Munich Re combines its long-standing “expertise in many domains, and various lines of business” with the technological pillar of their AI expertise. To leverage the wealth of experience even better, Munich Re is continuing to focus on qualification and upskilling.
Our data science and data analytics teams are thriving all over the world. On top, we provide a large amount of colleagues with very specific training.
— Andreas Bayerstadler, Head of Central Analytics, Munich Re
The expertise on the Munich Re staff is profoundly deep; irrespective of whether they bring with a background in natural science, math or physics, the specifically designed training regimen enabled them to expand this level of expertise. “We invented a citizen data science program. Our black belt program,” Bayerstadler said.
Bayerstadler adds, employees without technical backgrounds should also get the opportunity for upskilling. “We’re about to launch a green belt program,” he said. “We want to motivate them even if some may are a little afraid of formulas. Dataiku is a big part of that.”
A Shift in Mentality
For Bayerstadler, the core of constant internal cross-skilling is to change the mindset from heavy data management to one of advanced analytics. “Our underwriters are using and managing data from multiple clients in various formats. And some are struggling with this. We intend to take them on a journey to more advanced analytics, starting with simpler things likeclustering, then deep learning methods, LP methods, andGenerative AI,” he explained. The goal is, beyond automation and efficiency gains, to make better data-driven decisions.
Dataiku Helps Munich Re Build Technical Knowledge
Dataiku’s solutions helped Munich Re get comfortable with designing and using Generative AI. Bayerstadler: “It starts with a data literacy program for everyone in the company.”
With Dataiku, the technology gets much simpler. It means technical engagement and working with data in practice.
— Andreas Bayerstadler, Head of Central Analytics, Munich Re
Dataiku continues to support Munich Re’s existing black belts through customized, individualized courses. The experienced set of users is comfortable with coding and Python to take advantage of more complex AI and ML configurations. To implement similar solutions for the green belts, coding elements are to be removed.
“The cool thing is, we already have quite a few AI-based data management features with plugins in our workflow, and offer — through Dataiku — a frontend where people can really leverage the whole range of analytical work,” he explained. This difference is key, especially for these green belts.
Going forward, we would like to host and train our own language models and make them available through Dataiku.
— Andreas Bayerstadler, Head of Central Analytics, Munich Re
Looking Ahead to Deeper Democratization
Bayerstadler stresses the importance of continuous training to keep skills sharp over the long term.
For Bayerstadler, communication throughout the process is also key to success. “We need to involve people, talk to potential users, identify valuable use cases, then showcase the value of the tool, of AI as a topic, to then scale and leverage it into an AI driven business.” Overall, Munich Re’s approach to increasing the availability and ease of use is a clever combination of in-house experience and expertise combined with available technical AI knowledge.
Share
Malakoff Humanis: Improving Customer Relations With the Power of NLP
To address their growing challenges in keeping up with customer demands and providing quality customer service, Malakoff Humanis turned to Dataiku’s Deep Belief program and collaborated with Dataiku’s data scientists on two advanced natural language processing (NLP) projects.
Aviva has been using Dataiku for many years to build a range of data analytics and machine learning use cases, ultimately forwarding their customer-first mission.
OHRA built a robust, integrated ecosystem within Dataiku that allows them to test, build, and deploy data solutions 6x faster, while also ensuring customers are treated fairly.
Each of Heraeus’s 20 operating companies has its own lead identification and qualification process. See how Heraeus uses Large Language Models (LLMs) in Dataiku to support these processes, ultimately saving time and increasing sales conversion.
