fr
en
Operationalise and manage data projects with MLOps
An AI model only delivers lasting value when its operation is well managed. With MLOps, connect deployment, monitoring and maintenance to organise the transition to production. Strengthen your ability to coordinate data and operations teams around models monitored over time.
- Duration
- 2 days 14 hours
- Code
- IA030FR Code
Presentation
Moving from experimentation to production remains a major challenge for artificial intelligence projects. Many initiatives fail because production processes are insufficiently robust and teams do not collaborate smoothly. This 2-day course provides the methodological and technical foundations of MLOps (Machine Learning Operations) to make the entire model lifecycle reliable, from design to operational maintenance.
The programme covers the whole value chain: structure automated data pipelines, implement continuous integration and continuous deployment (CI/CD), and ensure production model quality. Beyond technology, it emphasises data governance and the alignment required between Data Scientists, DevOps teams and business stakeholders for project success.
Through immersive practical workshops, you will use industry-standard tools and simulate real deployment and management situations. You will leave with a clear view of how to establish an agile approach suited to data, anticipate performance drift and ensure your AI solutions' ethical and regulatory compliance.
Objectives
By the end of this MLOps course, you will be able to:
- define MLOps strategic and operational challenges within the AI project lifecycle;
- structure key data project stages, from initial exploration to production monitoring;
- orchestrate effective agile collaboration between data, IT and business professionals;
- deploy robust MLOps architectures incorporating versioning and CI/CD automation;
- manage governance, ethics and model performance risks over time.
Program
Module 1: understand MLOps fundamentals and project framing
- Defining key concepts and distinguishing DevOps and MLOps approaches.
- Analysing the specific challenges of operationalising AI projects.
- Identifying business needs and defining success indicators (KPIs).
- Building a multidisciplinary project team suited to data challenges.
Practical exercises
- Map an AI model's complete lifecycle within a typical organisation.
- Develop a structured project brief to launch a fictional data project.
Module 2: master technical architecture and pipelines
- Designing automated data processing and model training pipelines.
- Implementing dataset and model versioning.
- Applying continuous integration (CI) and continuous deployment (CD) principles to machine learning.
Practical exercise
- Build a simplified MLOps pipeline using a leading open-source tool.
Module 3: manage production deployment and monitoring
- Strategies for deploying models to production and making them accessible to users.
- Performance monitoring and data drift detection.
- Lifecycle management: retraining, updating and retiring obsolete models.
Practical exercise
- Implement a monitoring dashboard to track model health in real time.
Module 4: ensure project governance and agility
- Secure access management and sensitive data protection.
- Ensuring regulatory compliance (GDPR, AI Act) and algorithm auditability.
- Adapting agile methods to the uncertainty inherent in data science projects.
- Using collaborative tools to streamline technical-business team communication.
Practical exercises
- Analyse ethical and regulatory risks in a concrete use case.
- Simulate an agile sprint for a data project.
Audience
This course is intended for technical and business stakeholders in data projects, including:
- data and AI project managers orchestrating delivery and ensuring deadlines are met;
- AI product owners defining product vision and prioritising high-value features;
- data scientists and data engineers seeking to professionalise production deployment methods;
- technical architects designing infrastructure supporting data pipelines;
- innovation managers leading digital transformation and business data strategy.
Prerequisites
This course has the following prerequisites:
- Professional experience: practical experience in technical or project management environments is recommended to understand production-readiness concepts.
- Basic knowledge:
- understanding of machine learning fundamentals;
- familiarity with cloud environments or DevOps culture is an advantage.
Teaching and assessment methods
- Initial skills assessment
- Training materials provided to participants
- Continuous assessment throughout the course
- End-of-course feedback questionnaire
- Combination of theory and practical application
- Attendance records
- Post-course follow-up evaluation
- Practical exercises
Course highlights
- A cross-functional perspective: acquire combined technical and methodological skills essential for connecting business and IT teams.
- Practical, tool-based learning: use concrete CI/CD and monitoring solutions for immediate operational readiness.
- Governance focus: integrate ethics and regulation from the design stage to safeguard projects.
- Active learning: consolidate knowledge through 6 practical workshops covering the full project lifecycle.
Dates and sessions
Choose the date and delivery format that suit you.
No upcoming sessions are currently available.
Any brand names and logos mentioned in this course description (e.g. MLflow, Jira, Trello) belong to their respective owners. Their educational use does not constitute an endorsement or partnership.