Work Package 5 focuses on advancing mobile, micro-credential-based learning in smart farming through research, best-practice development, pilot implementation, and global dissemination. Activities include analysing existing mobile learning tools and platforms, benchmarking quality standards, and designing pilot courses that integrate interactive, mobile-friendly content. The original mobile learning concept evolved into an AI-powered LLM Tutor approach, enabling personalized, adaptive, and scalable learning experiences. Results will be shared globally—particularly in low-income countries—through partnerships with ENAMA and UN-CSAM, aiming to equip farmers with digital skills to address climate change, reduce energy use, and strengthen food security. A best-practice guide will ensure quality, accessibility, and flexibility in bridging education and professional needs.
Activity 5.1 – Research and Comparative Analysis
Research on existing mobile tools, along with analysis and adaptation for online and literature, will be conducted. Additionally, BOKU will provide research on existing micro-credentials for mobile learning. This activity will involve checking the e-learning industry’s list of the best tools for finding mobile learning platforms, exploring Learning Management Systems (e.g., Moodle, Canvas, Sakai, Open edx, TalentLMS, Adobe Captivate Prime LMS, Docebo) for on-the-go learning features, evaluating mobile learning platforms and user reviews, and comparing LMS solutions to find the right options. The research also investigates m-learning technologies (e.g., Ad-Connect, Kolibri, Rumie, Eneza Education) and analyses FAO Virtual Learning Centres as an example of good practice.
Activity 5.2 – Best Practice Guide for a mobile, micro-credential-based learning environment
Boku establishes benchmarks such as institutional support, development of micro-credentials, teaching and learning, teaching methods and assessment, learner support, faculty support, etc., for best practices to guarantee the success and quality of education in a mobile, micro-credential-based learning environment. BOKU will outline the guide’s structure and author the content, soliciting contributions from partners. Findings from the learning platform research are being compiled into a Best-Practice Guide, giving special consideration to FAO Virtual Learning Centres.
Activity 5.3 – Implementation of pilot courses for mobile phones and mobile learning
BOKU will collaborate with all partners to develop mobile learning courses, integrating all pertinent information derived from comprehensive analysis. All partners will contribute to the learning design for effective mobile learning and digital experiences, update content resources, and incorporate social and interactive elements. They will also take into account the user experience to motivate the target groups and to increase effectiveness.
The original concept of implementing mobile learning courses —focused on mobile-friendly content—has evolved into the more advanced LLM Tutor approach. While the initial goal emphasized delivering digital education via mobile platforms, the shift toward integrating Large Language Models (LLMs) introduces a dynamic, AI-powered tutoring system. This new model not only presents course content but also personalizes learning by adapting explanations to the learner’s level, answering questions interactively, and offering feedback. The LLM Tutor thus enhances the original mobile learning idea by enabling deeper engagement, scalability, and learner-centered support through AI.
The USAGE-NG project is an Erasmus+ initiative modernising agricultural engineering education for small and medium farms in the face of climate change. It emphasizes a micro-credentials approach and mobile learning to upskill farmers and students through flexible, digital courses. Work Package 5 (WP5) specifically explored mobile didactics and learning platforms, aiming to deliver short courses via smartphones to reach geographically dispersed learners. Within this context, Activity 5.3 piloted an AI-driven tutor as an innovative approach to mobile learning. This report presents: (1) the rationale for shifting from traditional mobile courses to an AI tutor, (2) the conceptual design of the AI tutor in USAGE-NG, (3) a step-by-step workflow for building the AI tutor, and (4) an explanation of the AI tutor’s system prompt structure. The aim is to inform EU education and rural development policymakers about the outcomes and lessons of Activity 5.3 in supporting lifelong learning for farmers.
Within the USAGE-NG project, a structured workflow was developed to prepare existing university teaching materials so that a Large Language Model (LLM), such as ChatGPT, Mistral, or a locally hosted model, can function as a digital tutor for course content. The workflow focuses on pedagogical preparation rather than technical complexity and includes collecting all relevant materials, creating structured summaries, transcribing lecture recordings, and organising exercises in a consistent and transparent way. These materials are embedded into a customised Artificial Intelligence (AI) system, resulting in a chatbot that can explain concepts, answer questions, and respond flexibly to different learner needs based on curated course content.
The AI Tutor is designed to act as a supportive explanatory layer rather than an authoritative instructor. By grounding responses in validated academic materials and clearly defined system prompts, the tutor can adapt explanations to learners’ prior knowledge, provide just-in-time clarification, and integrate insights from lecture transcripts that are often absent from static slides. This dialogical approach addresses common challenges in University Lifelong Learning, such as heterogeneous learner backgrounds, limited time availability, and the need for flexible, asynchronous support, while preserving alignment with institutional curricula and learning objectives.
Importantly, the workflow emphasises responsible and sustainable use of AI in education. Educational value is understood to arise not from automation alone, but from careful pedagogical design, transparent use of sources, and clear boundaries between human and artificial roles. Legal, ethical, and organisational considerations, such as copyright restrictions, data protection, and institutional governance, are treated as integral to the implementation process. Positioned in this way, the AI Tutor complements existing teaching practices and supports core principles of University Lifelong Learning: learner- centredness, flexibility, inclusivity, and long-term sustainability within established university structures.
The following demo files illustrate how existing university courses were prepared and structured for integration into an AI-supported tutoring environment within the USAGE-NG project. Each set of materials demonstrates the practical workflow described in the AI Tutor manual, including structured summaries, transcript integration, exercise preparation, and system prompt design. The demos serve as concrete examples for educators who wish to replicate this approach in their own teaching contexts.
GIS Course
The GIS demo files showcase how a train-the-trainer course in Geographic Information Systems was transformed into an AI-supported learning environment. Materials include structured lecture summaries, integrated transcript insights, and practice exercises related to mapping, spatial analysis, and applied fieldwork. The demo illustrates how technical content can be reorganised to enable conversational explanations and just-in-time learner support, particularly in resource-constrained or mobile learning contexts.
Smart Agriculture Technologies for Mountain Ecosystems
This demo builds on a hybrid course focusing on smart technologies tailored to small-scale and mountain farming systems. The files demonstrate how laboratory exercises, field demonstrations, and theoretical inputs were systematically structured for AI integration. Particular emphasis is placed on adapting explanations to heterogeneous prior knowledge and embedding contextual examples relevant to alpine agriculture and smallholder environments.
Smart Farming and IoT
The Smart Farming and IoT demo files present a technology-focused module covering digital farming tools, sensors, GNSS, data management systems, and Internet of Things (IoT) applications. The materials illustrate how complex engineering concepts were summarised, enriched with transcript-based clarifications, and converted into interactive AI-supported exercises. The demo highlights the tutor’s ability to adjust explanations for beginner and advanced learners, supporting flexible lifelong learning pathways.
Together, these demo files provide practical templates for transforming existing course materials into structured, AI-ready knowledge bases that enhance accessibility, adaptability, and pedagogical coherence in university lifelong learning.
Activity 5.4 – Mobile Course dissemination to low-income countries
Dissemination efforts will be directed towards local, national, and international private firms, non-governmental organizations, contractors’ associations, as well as national and international representatives of the agricultural sector, engineering firms, and professional associations. The USAGE-NG project has thoroughly investigated the smart farming education needs, particularly focusing on small/medium-hold farmers, and has identified several key areas and strategies for addressing these needs. Dissemination Events where held throughout the project. There will be a final dissemination conference.