Applied Machine Learning & Artificial Intelligence

Transfer of cutting-edge research to individual industry problems

Contact

image/svg+xml Cloud Computing ML & AI A P I

We are funded by the Federal Ministry for Economic Affairs and Energy

and the European Social Fund in the context of the EXIST stipend.

WUI Wildfire risk modeling with Artificial Intelligence

Dynamic risk modeling using remote sensing data


  • Our solution closes the information gap for risk modeling of wildland–urban interface (WUI) areas. WUI areas are a fast expanding risk pool for insurer.

  • In 2017 alone, US wildfire damage amounted to $16.2 billion. In comparison, in the 10 year period before (2007-2016) combined losses totaled roughly $15.8 billion.

  • Risk models based on claim history have limited predictive power for WUI areas. We provide additional data sources to improve risk model accuracy.


Our Solution


Our AI-models automatically detect wildfires on satellite images and digitize & categorize this information combined with additional data sources.


Our dynamically updated risk database contains data with a very high temporal & spacial resolution worldwide.


We provide a high throughput and low latency API for accurate and continuous risk assessment.


Digitalization & Automation

Sustainable digitalization and process automation using Artificial Intelligence (AI) and Machine Learning (ML) methods.

Data Digitization

  • Our machine learning models intelligently extract & categorize information from satellite images, health records, PDFs, location data, identity cards and more.
  • We use state-of-the art techniques in image recognition, natural language processing and signal processing to extract & categorize information.

Process Automation

  • We automate processes with ML that otherwise require human supervision.
  • We work on applications ranging from predictive maintenance in manufacturing to underwriting insurance risks.
  • We transfer existing processes into formal specifications that can be solved by ML-methods.

Prototyping

  • We believe that ML-solutions are best developed with iterative end-to-end prototypes.
  • This allows us to ensure that data quality & availability and model complexity reflect the business objective.
  • We develop prototypes that fulfill all critical requirements of the production system.
Our experienced team members worked in digitalization & research projects with:
Bayer Technology Services - Google Research - Victorinox - Swiss Re - IFH Cologne

Custom Machine Learning Solutions

Transferring academic & industry machine learning research into powerful solutions requires a combination of deep technical understanding, state-of-the-art tools, and the right development approach.

Artificial Intelligence & Machine Learning

We transfer cutting-edge academic & industrial research results in ML to industry applications.
  • ML research experience with industry leaders (Bayer AG, Swiss Re, Google Research)
  • Developer of popular ML library (fastFM)
  • Winner of high profile ML-competitions
  • Publications in top tier ML journals & speaker on conferences (WWW, ECML, PAKDD, JMLR)

✔ Profit from our ML-experience & expertise

Cloud Computing & Data Engineering

We build on powerful cloud computing technologies allowing us to quickly create individual & complex ML solutions.
  • Build on proven open source libraries (Nomad, Kubernetes, Docker, etc.)
  • Handles > 7 TB data per day
  • Secure & quickly scalable
  • Cross-cloud deployments
  • Funded by Google Cloud for Startups
  • N.A

✔ Save time & costs with our advanced Tech Stack

Agile Data Science & Software Engineering

We transfer proven software development practices to data science projects to iterate at agile speed.
  • Early prototype & continuous improvements
  • Continuous integration & delivery
  • Peer-reviewed code & easily deployable solutions (docker container)
  • N.A
  • N.A
  • N.A

✔ Work with maximum efficiency at agile speed

Management

Sonja Strothmann
CEO

Sonja holds a Master of Science in Strategic Marketing from the Maastricht University. She has worked for +9 years on diverse Digitalization Projects as IT-Manager and Consultant for known Companies. At Victorinox in Switzerland she was responsable for the Online Operations Management.

Dr. Immanuel Bayer
CTO

Immanuel holds a Diploma in Computational Engineering Sciences from the RWTH Aachen and a PhD degree in Machine Learning & Computer Sciences from the Konstanz University. He worked at Bayer Technology Services, Google Research in Mountain View and Swiss Re in Zurich.

Drop us a line or two

We'd love to hear from you