Our Mission
Artificial intelligence is a core driver for innovations in almost all fields.
But only if done right.
Besides algorithmic excellence, successfull projects require experience in software engineering and a deep understanding of business needs.
That’s what we offer.
Combining research and practice, we don’t merely talk about AI-based systems.
We build them.
Features and Details
We integrate and transform data, build perfectly coordinated machine learning models, and transfer them into productive operation. Everything in one place, highly automated, and always under your control.
Data Adapters
We provide data adapters for all major data formats and databases. With only a little information about the data structure, we can integrate any data into our platform and process it fully automatically. All this in a scalable infrastructure designed to handle terabytes of data. Our integration process cleans up the data and ensures the necessary data quality for further processing.
Data Augmentation
Sometimes it takes a little more knowledge. Our platform allows to extend data automatically with additional information about weather, traffic, holidays, and more. We integrate different data providers in one place and remove the need for additional contracts and technical effort.
Prebuild Models
We already implemented all essential algorithms into our platform and can quickly select those you need. As our data integration process converts the data into a reliable quality and structure, your models can be trained without additional effort. Automated optimization processes ensure that each model is perfectly tuned. Already during training, the platform monitors the model to identify and eliminate classic pitfalls.
Zero Time Deployment
Our goal is to make your project operational. We don't just do data science -- we engineer complete AI applications. Our platform provides a large number of interfaces to forecasts tailored to your specific needs. And all our models are compatible, so productive operation is only one click away. No more costly follow-up projects necessary!
Quality Monitoring
We create confidence in AI with a broad evaluation suite to constantly monitor our models. The exact requirements may differ with your project and are hard to generalize. If you require a higher accuracy, just report back at any time. Our platform automatically translates your feedback into actions and identifies possible model adjustments, which you can accept just in time.
Continuous Optimization
In order to remain accurate, a machine learning model must adapt to the new data. Models running on our platform are automatically trained at short intervals to ensure the best possible performance at all times. If the available data changes, the models are revised in close cooperation with you. We check whether the current models are still suitable and examine additional possibilities to even further improve the prediction quality.
Our Team
Building AI-based systems requires excellence in cutting edge technologies and at the same time ambition for simple economic solutions. That’s why we’ve put both into our team.
Our Publications
Our work roots in science. And we still do science. We submit our methods, tools, and techniques to scientific venues and have them peer-reviewed by the scientific community. To make sure everything we do is rock solid. Here is a list of our recent work.
Towards a Software Engineering Process for Developing Data-Driven Applications Inproceedings Proceedings of the 7th International Workshop on Realizing Artificial Intelligence Synergies in Software Engineering, [email protected] 2019, Montreal,
QC, Canada, May 28, 2019, pp. 35–41, 2019. Towards Concept Based Software Engineering for Intelligent Agents Inproceedings Proceedings of the 7th International Workshop on Realizing Artificial Intelligence Synergies in Software Engineering, [email protected] 2019, Montreal,
QC, Canada, May 28, 2019, pp. 42–48, 2019. A Decentralized Architecture and Simple Consensus Algorithm for Autonomous Agents Inproceedings Proceedings of the 12th European Conference on Software Architecture:
Companion Proceedings, ECSA 2018, Madrid, Spain, September 24-28,
2018, pp. 7:1–7:4, 2018. Towards a Practical Process Model for Anomaly Detection Systems Inproceedings Proceedings of the 1st International Workshop on Software Engineering for Cognitive Services, [email protected] 2018, Gothenburg, Sweden, May
28-2, 2018, pp. 41–44, 2018. 2017 IEEE Congress on Evolutionary Computation, CEC 2017, Donostia,
San Sebastián, Spain, June 5-8, 2017, pp. 1908–1915, 2017.2019
2018
2017