Introduction

In this technical workshop, you will be introduced to the practice of MLOps — the domain where machine learning, software development and operations meet. You will learn how to develop, test, deploy and manage machine learning models according to robust, scalable workflows.

We cover the entire journey: from experiment tracking and model versioning to CI/CD, containerization, monitoring and retraining strategies. You will get hands-on with tools such as MLflow , DVC , Weights & Biases , Docker , Kubernetes and Kubeflow Pipelines .

This training provides a solid foundation for teams that want to seriously deploy machine learning in production environments and that want to build robustness, transparency and control into their AI processes.

Learning objectives

1. Insight into the complete MLOps process

• Lifecycle of an ML model: from data preprocessing to deployment and monitoring

• Structuring code and experiments for reproducibility

• Version control of models, datasets and pipelines

2. Working with modern MLOps tools

• MLflow for experiment tracking, model registration and model deployment

• DVC (Data Version Control) for dataset management and pipeline tracking

• Weights & Biases for advanced logging and visualization of training runs

• Kubeflow for orchestrating ML workflows in the cloud or on-premises

3. Deployment & CI/CD

• Setting up pipelines with Git , Docker , Kubernetes and GitHub Actions/GitLab CI

• Automate training and deployment with CI/CD workflows

• Deployment strategies: batch, real-time, edge, model serving via REST APIs

4. Monitoring, observability and model maintenance

• Setting up monitoring for model performance and data drift

• Identifying performance loss or bias in production

• Retraining strategies and MLOps loop (continuous improvement)

• Feedback loops with user data or annotators

Approach and working methods

This workshop is hands-on and focused on building a working MLOps pipeline. You will work with Python, Jupyter Notebooks and command-line tools in combination with cloud or local environments. The training is modular: depending on your team we can focus on specific components such as experiment tracking, CI/CD or monitoring.

Examples of applications:

• Automatic model rollout after new training rounds

• Continuous validation of predictions on live systems

• Structured collaboration in ML teams with shared workflows and tools

For whom

This course is intended for data scientists, ML engineers, MLOps specialists and DevOps professionals who want to professionally manage machine learning projects and bring them to production. Experience with Python, git and basic knowledge of ML models is recommended.

Interested in this training?

Feel free to contact us. We are happy to think along with you about a customized training, tailored to your infrastructure, team and phase of AI maturity.