FM2AZDM20 Deployment and Maintenance
FM2AZDM20 Deployment and Maintenance
- Course description
- Course CodeFM2AZDM20
- Level of Study5.2
- Program of StudyApplied Artificial Intelligence
- Credits20
- Study Plan CoordinatorBertram Haskins
This course provides students with the knowledge and practical skills needed to successfully deploy, monitor, and maintain AI/ML solutions in real-world environments. Students will learn version control practices, virtualisation technologies, and system architecture design for AI applications.
The course also covers key aspects of MLOps, including model drift detection, ongoing model maintenance, and effective team collaboration in AI projects. Additionally, students will gain experience in designing interactive dashboards and user-friendly outputs, ensuring that AI insights are accessible to diverse audiences. Ethical and legal considerations in AI deployment will be emphasised, along with hands-on experience in constructing project proposals and developing an exam project.
The candidate...
- has knowledge of concepts, theories, models, processes and tools used in the deployment, monitoring, and maintenance of AI/ML systems, including version control and collaboration platforms supporting the AI lifecycle.
- has knowledge of virtualisation and containerisation technologies for scalable and reproducible AI/ML model deployment.
- understands the principles of system and cloud architecture design supporting AI applications in production environments.
- is familiar with MLOps concepts and lifecycle models, including automation, continuous integration and delivery, and quality assurance of AI models.
- has knowledge of the causes, detection, and impact of model drift, and techniques for ongoing monitoring and mitigation.
- has knowledge of communication and visualisation methods, such as dashboards and interfaces, to make AI results accessible to diverse users.
- has insight into ethical and legal frameworks governing AI deployment, model maintenance, and data protection.
The candidate...
- can apply vocational knowledge to practical and theoretical problems in tracking, deploying, and maintaining AI/ML models using version control systems.
- masters relevant vocational tools, materials, techniques and styles, such as virtualisation and container orchestration tools for model deployment.
- can design and implement system and cloud architectures that support reliable and scalable AI/ML services.
- can apply agile and collaborative workflows to coordinate AI/ML development, deployment, and maintenance tasks.
- can study a situation and identify subject-related issues and what measures need to be implemented, such as streamlining or automating deployment and monitoring processes through MLOps.
- can explain his/her vocational choices when selecting model monitoring strategies or addressing drift mitigation in production environments.
- can develop user-oriented dashboards and visual interfaces to communicate AI insights effectively to both technical and non-technical audiences.
- can prepare and document a project or exam proposal, demonstrating the ability to integrate and apply vocational knowledge across deployment and maintenance contexts.
- can reflect over his/her own vocational practice and adjust it under supervision, for example refining deployment workflows or dashboard designs as part of an exam project.
- can find and refer to information and vocational material and assess its relevance when evaluating tools, frameworks, or deployment architectures.
The candidate...
- can plan and carry out vocational tasks and projects alone or as part of a group and in accordance with ethical requirements and principles, applying best practices in AI deployment, monitoring, and maintenance to ensure sustainability and efficiency.
- understands ethical and legal principles that regulate AI deployment and maintenance, including accountability, bias mitigation, and data privacy.
- can communicate technical and conceptual aspects of AI deployment clearly to stakeholders with varying technical backgrounds.
- can build relations with peers, cross-disciplinary partners, and external stakeholders, collaborating effectively within AI/ML deployment and maintenance projects.
- can continuously update his/her vocational knowledge and adapt to emerging tools and trends in AI deployment and maintenance, contributing to organisational development.
Digital Learning Resources
The learning management system (LMS) is the primary learning platform where students access most of their course materials. The content is presented in various formats, such as text, images, models, videos or podcasts. Each course follows a progression plan, designed to lead students through weekly modules at their own pace. Exercises and assignments (individual or in groups) are embedded throughout the courses to support continuous practice and assessment of the learning outcomes.
Campus Resources
In addition to the digital learning resources, campus students participate in physical learning activities led by teachers as part of the overall delivery.
Guidance
Guidance and feedback from teachers support students' learning journeys, and may be provided synchronously or asynchronously, individually or in groups, via text, video or in-person feedback.