This minor course addresses the growing professional demand for engineers who can design, implement, and responsibly deploy Large Language Models (LLMs) across applied domains such as healthcare, education, business innovation, and creative industries. Students learn to translate state-of-the-art LLM research into reliable, ethical, and production-ready applications that solve concrete problems, such as automating knowledge retrieval, building conversational assistants, or improving information access through generative interfaces.

Throughout the minor, students develop professional competencies in natural language processing, software engineering, and ethical AI. They create professional products that reflect real industry roles: LLM-based applications, prompt-engineered pipelines, evaluation reports, and responsible AI documentation. The capstone industry project integrates these outcomes into a complete AI solution for an external partner, demonstrating students’ readiness to operate as AI engineers, applied AI developers, or data science professionals in the rapidly evolving field of generative AI.

The Large Language Model Engineering minor is highly relevant for students in Software Engineering, Data Science, Artificial Intelligence, and related applied science programmes, as it builds directly on their existing programming and machine learning knowledge. By focusing on the design, fine-tuning, and deployment of LLMs, students strengthen professional competencies that are increasingly in demand across industries.

The minor directly connects to professional practice through its capstone project, where students collaborate with companies and organizations on real-world applications, ensuring immediate relevance to the labor market. The topicality is evident in the rapid adoption of generative AI across domains such as healthcare, education, business services, and creative industries. Where possible, the minor will align with research activities within the Hogeschool’s Research Centre Creating 010 (Digitization and social transformations).

Students acquire cutting-edge competences including prompt engineering, parameter-efficient fine-tuning, Retrieval-Augmented Generation, and ethical AI design. These skills prepare graduates for roles that require advanced AI expertise and the ability to translate innovations into applied solutions.

The program is inherently interdisciplinary: students combine knowledge from computer science, data science, software engineering, and ethics, while also integrating perspectives from industry partners in fields such as healthcare, finance, logistics, and education. This ensures students learn to apply LLM technologies across multiple disciplines, gaining both technical depth and cross-domain adaptability.

In addition, the learning outcomes and competences of this minor are explicitly derived from the national HBO frameworks for Informatica, and Applied data science & artificial intelligence. They extend the generic HBO competences, such as design and realization, ethical reflection, and professional communication, and contextualize them within the field of Large Language Model engineering. This ensures that the minor’s formulation of competences is consistent with national profiles while being specifically tailored to the emerging professional context of generative AI.

The formulation of competencies in this minor is therefore not new, but a translation and contextualization of the existing national profiles to the rapidly evolving AI domain, ensuring both national consistency and alignment with professional practice.

Content and program
The Large Language Model Engineering minor (30 EC) provides students with both theoretical foundations and practical experience in working with Large Language Models (LLMs).

The content is divided into interconnected modules, each focusing on a key competence area:

• Foundations of NLP and LLMs: Tokenization, embeddings, transformer architectures, attention mechanisms, and scaling laws.
• Ethics and Responsible AI: bias, fairness, privacy, security, transparency, and relevant regulatory frameworks (e.g., EU AI Act).
• Pythonic LLM Engineering: hands-on work with Hugging Face, LangChain, vector databases, parameter-efficient fine-tuning (LoRA, QLoRA), and evaluation frameworks.
• Applied LLM Systems: Retrieval-Augmented Generation (RAG), multi-agent systems, deployment strategies, and monitoring pipelines.
• Capstone Industry Project: Applied project in cooperation with industry or external partners where students deliver an end-to-end LLM solution. 

Balance of Practice and Theory through Contextualized Learning

The minor integrates theory and practice in parallel. Each theoretical lecture is reinforced by labs and workshops where students apply concepts immediately in code. Ethical principles are contextualized within technical exercises (e.g., dataset selection, fine-tuning, prompt engineering). The capstone project ensures authentic, contextualized learning by engaging students in solving real-world challenges provided by industry partners and other professional sectors. This balance guarantees that students not only understand the principles behind LLMs but can also apply them in professional practice at HBO level.

Learning Objectives
The Large Language Model Engineering minor equips students with professional competences at the higher professional education level, explicitly aligned with the Dublin descriptors.

Learning Objectives (LOs)

The following learning objectives are derived from the above professional competencies. By the end of the minor, students will be able to:

1. Explain the principles of natural language processing, transformer architectures, and large language models. 
2.  Design, implement, and deploy LLM-powered applications using state-of-the-art tools and frameworks (e.g., Hugging Face, LangChain, vector databases). 
3. Critically evaluate ethical, societal, and technical implications of LLMs, including bias, privacy, and safe deployment, and integrate responsible AI principles into system design.
4. Collaborate in multidisciplinary teams and effectively communicate technical designs, ethical assessments, and project outcomes to both technical and non-technical stakeholders. 
5. Independently acquire new knowledge in the rapidly evolving field of AI and apply it to innovative professional contexts, demonstrated in the capstone industry project.

1. Target Audience
The minor is intended for students interested in applied artificial intelligence, natural language processing, and software engineering. It is especially relevant for those who wish to gain practical expertise in building, fine-tuning, and deploying Large Language Models (LLMs), and for students who are motivated to apply these technologies in real-world professional contexts such as healthcare, education, logistics, or creative industries.

2. Eligible Study Programs
The minor is open to external and internal students from bachelor programs in:

• Software Engineering
• Artificial Intelligence
• Data Science
• Computer Science
• Information Technology
• Or other closely related applied science programs.

3. Entry Requirements
Students must have:

• Successfully completed at least 120 ECTS of their bachelor program (third-year level or higher).
• Fundamental programming knowledge, including proficiency in Python.
• Basic familiarity with algorithms, statistics, and data handling.
• Sufficient English proficiency to follow technical lectures and literature.

4. Additional Supports for students with less prior knowledge or programming experience
To accommodate differences in prior knowledge and programming experience, the minor provides differentiated learning support. During the first two weeks, students complete short diagnostic coding tasks to identify their individual skill levels. Those who need reinforcement receive guided practice materials and mentoring during lab hours, while more advanced students are encouraged to take on extended challenges or contribute to peer support. This ensures that all participants can engage effectively with the technical components of the course regardless of their background.

5. Capacity and Enrollment Policy

• A maximum of 30 students will be admitted per intake.
• Applicants are required to submit a short motivation letter explaining their interest in and motivation for enrolling in this course.
• Motivation letters will be reviewed to assess applicants’ suitability and commitment to the course. If, after this review, the number of eligible applicants exceeds the available capacity, students will be admitted on a first-come, first-served basis according to their registration date.

The prescribed literature supports both the technical learning objectives (foundations, engineering methods, applied systems) and the ethical objectives (responsible AI, bias, privacy, regulation).

1.     Avinash Manure, Shaleen Bengani, Saravanan S, 2022. Introduction to Responsible AI: Implement Ethical AI Using Python. Apress.
https://www.oreilly.com/library/view/introduction-to-responsible/9781484299821

2.     Bill Franks, 2020. 97 Things About Ethics Everyone in Data Science Should Know. O'Reilly Media, Inc.
https://learning.oreilly.com/library/view/97-things-about/9781492072652/

3.     Daniel Jurafsky and James H. Martin, 2025. Speech and Language Processing: An Introduction to Natural Language Processing, Computational Linguistics, and Speech Recognition, with Language Models. Freely available on author’s webpage:
https://web.stanford.edu/~jurafsky/slp3/

4.     Lewis Tunstall, Leandro von Werra, Thomas Wolf, 2022. Natural language processing with transformers. O'Reilly Media, Inc.
https://learning.oreilly.com/library/view/natural-language-processing/9781098136789/

5.     Lazy Programmer, 2024. Natural Language Processing - Transformers with Hugging Face. O'Reilly Media, Inc.
https://learning.oreilly.com/course/natural-language-processing/9781836200673/

6.     Denis Rothman, 2021. Transformers for Natural Language Processing. Packt Publishing.
https://learning.oreilly.com/library/view/transformers-for-natural/9781800565791/

This course is composed of 20 weeks, with each week featuring 2 days of face-to-face supervised learning. The remaining days of each week are allocated for self-directed learning, project activities, practicums, and independent study. The course will progressively cover fundamental through advanced topics in LLM engineering, with specific weekly topics and materials to be announced at the beginning of the semester.

– Individual Weekly Assignments – Weekly technical coding tasks providing continuous feedback to strengthen technical understanding and prepare students for summative assessments. 

– Capstone Project Proposal – Team submission including project problem, system architecture, and ethical considerations. Provides structured formative feedback before the final project. 

– Supervision & Peer Feedback – Continuous participation in supervision meetings and peer-feedback sessions to support reflective learning and collaboration.

– Final Exam – Written or online exam assessing theoretical and ethical understanding of NLP, LLM architectures, and responsible AI frameworks. Weight: 20%. Minimum passing grade 65%.

– Ethics Case Study – Analytical written assignment addressing ethical and societal challenges in LLM deployment. Weight: 15%. Minimum passing grade 65%.

– Capstone Project – Comprehensive industry-linked project delivering an end-to-end LLM application with documentation, evaluation, and ethical assessment. Weight: 60%. Minimum passing grade 65%.

– Final Presentation & Defense – Oral defense of the capstone project before instructors and/or industry partners, demonstrating integration, communication, and reflection. Weight: 5%. Minimum passing grade 65%; Mandatory participation.

Minor Choice Week | 2–5 March 2026

This minor does not participate in the Minor Choice Week 2026

--------------------------------------------------------------------------------------------

APPLICATION PROCESS (KOM)

Are you a student from another educational institution and would you like to follow a minor at Rotterdam University of Applied Sciences? If so, you must apply in two steps:

Step 1

• Register for the minor of your choice via the blue Apply button. You can find this button at the top right of the minor’s page.
• Download the learning agreement and complete it.
• Submit this learning agreement to the examination board of your study programme.

Once the examination board has granted approval, register for the minor in Step 2 no later than 01-07-2025 at 9:00 a.m.

Step 2
After approval, register via OSIRIS Application of Rotterdam University of Applied Sciences using the link below (first create an account).
https://osiris.hr.nl/osiris_aanmeld_hrprd/Welkom.do?proces=KOM2609&opleiding=MINOR-CMI-VT 00

Part of the application process is uploading the following documents:

• The learning agreement, signed by you and by your institution;
• A scan or photo of your passport or ID card.

You will be informed by Rotterdam University of Applied Sciences whether your application has been approved.

In OSIRIS Application, you must also upload the Proof of Paid Tuition Fee (BBC) for the academic year in which you wish to follow the minor. This can be done from 01 May 2026 onwards. You can request the BBC from your institution after you have signed or issued an authorization for the payment of the tuition fee for the relevant academic year. You may also choose the option for your institution to send the BBC directly to collegegeld@hr.nl.

You will receive a notification from Rotterdam University of Applied Sciences once your application has been approved.