About the school
This school aims to familiarize participants with the principles, architectures, engineering practices, and emerging research challenges in developing and maintaining Agentic AI systems. It emphasizes advancing from standalone LLM applications to systems featuring autonomous, collaborative AI agents capable of reasoning, planning, tool use, memory management, and adaptation in dynamic environments.
This intensive one-week program covers the foundations and engineering of Agentic AI from a systems perspective, blending lectures, research discussions, case studies, and practical exercises.
Topics include engineering foundations, LLMs, Generative AI, prompting, AI agents, multi-agent systems, design patterns, memory architectures, tool integration, orchestration frameworks, retrieval-augmented generation (RAG), Model Context Protocol (MCP), Agent-to-Agent (A2A) communication, software architecture, evaluation methods, and monitoring.
Each day features three sessions: two focused on theory, architecture, challenges, and research, and a final hands-on session where participants develop and test agentic workflows using modern frameworks. Hands-on activities involve building single-agent systems, orchestrating multi-agent workflows, integrating tools and memory, and assessing agent behaviors.
Program
Browse the program to learn more about the lectures and activities that will guide participants through a rich and engaging learning experience.
Foundations of AI Systems
- Foundations of Engineering AI Systems
- Large Language Models (LLMs) and Generative AI
- Prompting strategies and techniques
- Hands-on: Building your first single-agent system
Day 1 — Monday 14 September
AI Agents and Multi-Agent Systems
- AI agents: concepts and architectures
- Multi-agent systems and coordination
- Design patterns for agentic systems
- Memory architectures
- Hands-on: Orchestrating multi-agent workflows
Day 2 — Tuesday 15 September
Tool Integration and RAG
- Tool integration and external APIs
- Retrieval-Augmented Generation (RAG)
- Model Context Protocol (MCP)
- Hands-on: Integrating tools and memory into agents
Day 3 — Wednesday 16 September
Architecture and Orchestration
- Software architecture for Agentic AI systems
- Orchestration frameworks
- Agent-to-Agent (A2A) communication
- Observability and monitoring
- Hands-on: Building multi-tool orchestrated workflows
Day 4 — Thursday 17 September
Evaluation and Project Presentations
- Evaluation methodologies for agentic systems
- Assessing agent behavior and quality
- Final mini-project presentations
- Certificate distributions
Day 5 — Friday 18 September
Speakers
Assoc. Prof. Karthik Vaidhyanathan
Associate Professor,
IIIT Hyderabad
(Guest Speaker)
Prof. Valentina Lenarduzzi
Full Professor,
University of Southern Denmark
ACP Section Co-Chair,
Group Co-Head
Prof. Davide Taibi
Full Professor,
University of Southern Denmark
LEGO® Chair,
Group Co-Head
Asst. Prof. Matteo Esposito
Assistant Professor,
University of Southern Denmark
ECTS Credits
Earn ECTS credits upon successful completion of the school
1 ECTS credit
Lighter commitment
- Final project presentation only.
- Present your Agentic AI application at the end of the school week
5 ECTS credit
Full commitment
- Full written report and complete implemented project.
- Submission deadline: 30 days after the course ends.
Evaluation method
- Design and demonstrate an Agentic AI application
- Address a realistic problem scenario
- Assessed on system design and architecture
- Implementation quality and presentation
- Graded on pass / fail basis
Registration
Register before 30th July 2026
What is included
- Admission to all lectures and sessions
- Slides and materials from all lecturers
- Lunches on every day
- Coffee breaks throughout the day
- Social functions and networking events
- Access to hands-on lab sessions
Student
Student VAT-free rate.
3500
DKK
Academic Researcher
VAT-free rate for academic researchers.
4500
DKK
Industry
Practitioner and Industrial Researcher rate .
4500 + VAT
DKK
Frequently Asked Questions
Who can attend?
PhD students, postdoctoral researchers, academic researchers, and industrial professionals with an interest in Agentic AI systems. Basic programming knowledge is required.
What are the student registration fees?
For PhD students, proof of enrolment required. Fees include all lectures, materials, luncheons, coffee breaks, and social functions.
What is the registration deadline?
The deadline for registration is 9 August 2026.
Can I obtain ECTS credits?
Yes. 1 ECTS with final project presentation, or 5 ECTS with full report and implemented project submitted within 30 days after the course ends.
How is the evaluation conducted?
Participants design and demonstrate an Agentic AI application or workflow. Evaluated on system design, architectural reasoning, implementation quality, and presentation. Pass/fail basis.
What is the language of instruction?
All lectures, exercises, and materials are delivered in English.
What programming experience do I need?
Basic programming knowledge is required. Experience with Python is recommended as most hands-on sessions use Python-based agentic frameworks.
Who do I contact for more information?
Contact Prof. Davide Taibi at taibi@imada.sdu.dk