In addition to the pure provision of computing and storage resources, the AI service centre also offers support in the automation and optimisation of machine learning pipelines, as well as in the selection and training of models.
The services offered by the KI Service Centre are divided into three areas
Each of the areas presented helps to enable you to use artificial intelligence in your everyday life without barriers.
If you do not find what you are looking for in the offers below, please do not hesitate to contact our experts and work with them to define a suitable workshop or project.
Our cluster workshops help you to make the best possible use of our existing infrastructure and to develop and train your models efficiently in our Machine Learning Development Environment. We introduce you to the existing interfaces in detail and offer you the opportunity to apply what you have learnt directly in our test environment.
In this half-hour workshop, you will get to know the HPC Cluster from hessian.AI | The Hessian Centre for Artificial Intelligence and train a first example model on the cluster. We start with an introduction to the general operation of the cluster, the access options and the various interfaces. Each user then transfers data to the cluster independently, modifies a simple neural network and then trains it using several GPUs. We are happy to address your specific application questions.
In this half-hour workshop, you will get a rough overview of the structure and use of our cluster, familiarise yourself with the Keras interface of our HPC cluster and train models of varying complexity using the Keras interface on the cluster. We will be happy to address your specific application questions.
In this half-hour workshop, you will get a rough overview of the structure and use of our cluster and familiarise yourself with the PyTorch interface of our HPC cluster, train models of varying complexity using the PyTorch interface on the cluster. We will be happy to address your specific application questions.
In this half-hour workshop, you will get a rough overview of the structure and use of our cluster and familiarise yourself with the DeepSpeed interface of our HPC cluster, train models of varying complexity using the DeepSpeed interface on the cluster. We will be happy to address your specific application questions.
In this half-hour workshop, you will get a rough overview of the structure and use of our cluster and learn about the possibilities for hyperparameter tuning on our cluster. As part of the workshop, you will carry out a simple hyperparameter training session yourself using an example model. We will be happy to address your specific application questions.
The independent realisation of the first AI projects can present every user with various challenges and burn up a lot of resources and financial means in a short space of time. As part of our project support, we offer you the opportunity to first validate your ideas in a cost- and time-efficient manner and then turn them into a usable application as part of our small project.
Sometimes an idea sounds great on paper, but when it comes to practical implementation, challenges suddenly arise. A proof-of-concept (PoC) sprint is a quick and cost-effective way to find out whether a solution concept really works and to recognise potential problems at an early stage. This not only makes planning for the actual project easier, but also ensures that the time and money invested is not wasted in the end. Our PoC sprint is like a safe starting point for the successful realisation of your project.
As part of the AI Service Centre’s research, a large number of efficient, applicable models are developed and made available. In this workshop, you will use the expertise of our service staff to optimise one of these models for your use case and test it on our HPC computing infrastructure. Depending on the scope of the necessary optimisations, this results in different workshop durations.
Based on your proof of concept, our experts will support you in the development, implementation and training of your models. The scope of the project is customised to your requirements so that you can quickly put your idea into practice with our help.
The application and fine-tuning of existing models, as well as the adaptation of existing solutions to given circumstances, cannot always take into account the complexity and possibilities of artificial intelligence. Comprehensive further developments and research into new architectures offer unimagined possibilities, but on the other hand require extensive knowledge and comprehensive resources.
Our experts from the service centre and the hessian.AI ecosystem support you cooperatively in the implementation of medium and large AI projects and in the joint creation of new state-of-the-art solutions. Please get in touch with us via our contact form or at .
