DevOps for Data Science: 3-Wk Implementation
Kapernikov
Deploy data science for process optimization
Minimizing scrap, reducing raw materials or energy consumption, increasing production speed: these are some of the needs that you could face in your industrial process. Data Science can be used to gain actionable insights that support optimization of your process. Your process engineers know their processes best. So, we help them to integrate this data science research into your process in a robust way using the Azure cloud. As an end result, you will have tools put into place that visualize complex process insights. As a lot of AI projects fail in the deployment phase, we make sure that you get a robust and automated deployment process in your Azure cloud where user interaction is reduced to a minimum.
The solution is implemented using Azure DevOps and Azure ML. It is composed of two parts, a production pipeline and a development environment. The latter includes a template repository tailored for your data science project that is ready to make use of the production pipeline.
Production pipeline components
- Test, build and release of the code in Azure DevOps
- Training, testing and reporting of the models in Azure ML
- Artifacts storage using Azure DevOps
- Data and models versioning using Azure ML
- Deployment to Azure AKS is possible as an extra.
Advantages
This production pipeline frees up time for your process engineers to focus on what matters for your company: improving your industrial process. With this pipeline implemented, you can rest assured that if you have an urgent problem, you can react immediately, without sacrificing stability.
## Steps
- We define the technical requirements for the integration with the Azure cloud
- We configure the development environment and the production pipeline in Azure to integrate with your industrial process
- We test the full pipeline with you and we make you ready to use it in your data science project.