AI and machine learning
Our philosophy for ML/AI encompasses both the exploratory stages of model development as well as the processes needed to bring trained models to production services. This is why we advocate solutions that are built ground up to support flexible experimentation, high-performance training, and to let continuous delivery meet analytics. We specialize on open source software stacks and we are proponents of the Kubeflow initiative.
Cloud native computing
We are passionate about creating cloud native and vendor agnostic applications based on leading open source software stacks on top of Kubernetes, OpenShift and OpenStack. With our background in scientific computing, we know the importance of designing ground up for high performance and scalability. Automation is key to sustainable solutions, and this is why infrastructure as code is one of our core focus areas.
Data science & engineering
It all starts with data, and no machine learning project will succeed without sound principles for scientific data management. In fact, good data engineering practices is key to a unlocking long term business value from machine learning. Our team is highly experienced in solutions for secure, fast and highly scalable data analysis. We are specialized on connecting data science with the DevOps culture, using automated and repeatable data pipelines built on open source software stacks.
Scientific computing & HPC
Our experience in scientific computing spans the range from data-driven modeling to simulation to optimization. The fact that we have extensive experience with both traditional HPC and cloud computing makes us an ideal partner when developing performance critical applications leveraging cloud infrastructure, and when developing hybrid infrastructure tailored to meet both the needs for flexibility and performance posed by cloud native scientific computing applications.