Coordinators: Gilles MENEZ, maître de conférence and Johan MONTAGNAT, directeur de recherche, i3S laboratory (Université Côte d'Azur, CNRS)
FORMAT
Classroom
LOCATION
SophiaTech campus, Lucioles
PREREQUISITES
Coding experience with Python
CAPACITY
30 students
ABOUT THIS MINOR
- Summary
-
This course aims to provide an understanding of the principles and development of generative agent-based AI, with a strong emphasis on its practical use. It is illustrated through its application in the field of software development. It covers the history of AI development up to modern agent-based AI systems. It illustrates the principles of operation, the different types of models, and key settings through the installation and experimentation with various LLMs running on local resources, as well as complementary external tools. It draws on Anthropic’s Claude Code solution, which has been widely adopted by the professional community, to teach participants how to master AI-assisted software development. It covers the implementation of a RAG, the deployment of agents through an MCP interface, and multi-agent systems. It raises awareness of the risks associated with using AI models, their costs, their limitations, and the factors that can improve software development productivity.
LEARNING OUTCOMES
In this course, students will:
- Learn general principles of LLM and their different types;
- Distinguish between LLM, RAG, agent and agentic workflow;
- Deploy models on local resources to perform experimentations;
- Explore a professional agentic software development tool;
- Discover agentic AI capabilities in software development while acknowledging its limitations and proper use of such tools.
- Lecturers
-
- Mohamed KADRI, Monitoring and Observability Solution Architect at Sopra Steria – Field of expertise: Elastic Stack Architect, Software Engineering, Agentic AI
- Nikan Zandian JAZI, PhD student at CNRS, i3S laboratory – Field of expertise: spiking neural networks (SNNs), formal verification, large language models
- Bibliography
-
Scientific books:
- Alammar, J. & Grootendorst, M., Hands-On Large Language Models: Language Understanding and Generation, O'Reilly Media, 2024. ISBN 9781098150969
- Huyen, C., AI Engineering: Building Applications with Foundation Models, O'Reilly Media, 2025. ISBN 9781098166304
- Build a Large Language Model (from Scratch), Sebastian Raschka, Manning
Foundational articles available for self-service:
- Vaswani, A. et al., Attention Is All You Need, NeurIPS, 2017. arXiv:1706.03762
- Lewis, P. et al., Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks, NeurIPS, 2020. arXiv:2005.11401
- Liu, N. F. et al., Lost in the Middle: How Language Models Use Long Contexts, 2023. arXiv:2307.03172
- Becker, J., Rush, N., Barnes, E. & Rein, D., Measuring the Impact of Early-2025 AI on Experienced Open-Source Developer Productivity, METR, 2025. arXiv:2507.09089
Self-service technical resources:
- Boonstra, L., Prompt Engineering, Google / Kaggle whitepaper, 2024 — kaggle.com/whitepaper-prompt-engineering
- Google / Kaggle, Foundational Large Language Models & Text Generation — kaggle.com/whitepaper-foundational-llm-and-text-generation
- Wiesinger, J., Marlow, P. & Vuskovic, V., Agents, Google / Kaggle whitepaper — kaggle.com/whitepaper-agents
- Schluntz, E. & Zhang, B., Building Effective Agents, Anthropic, December 2024 — anthropic.com/research/building-effective-agents
- OpenAI, A Practical Guide to Building Agents, 2025 — A Practical Guide to Building Agents
- Boonstra, L., Spec-Driven Production Grade Development in the Age of Vibe Coding, Google / Kaggle whitepaper
- Anthropic, Prompt Engineering Guide — docs.anthropic.com (Build with Claude)
- Anthropic courses: Claude 101, ClaudeCode 101, Building with the Claude API, Introduction to Model Context Protocol
- Model Context Protocol specification — modelcontextprotocol.io
Security frameworks:
- OWASP GenAI Security Project, OWASP Top 10 for LLM Applications — genai.owasp.org
- OWASP GenAI Security Project, OWASP Top 10 for Agentic Applications — genai.owasp.org
- Evaluation
-
Type of evaluation Date or Submission deadline Time Location
(for on-site exams)% of the final grade Deliverable Submission deadline: 19/11/2026 N/A SophiaTech campus, Lucioles 1/3 Oral project presentation Date of the exam: 03/12/2026 15 minutes / presentation SophiaTech campus, Lucioles 2/3 Presence Presence is mandatory. A 1-point penalty will be applied to the final grade for each unjustified absence, up to a maximum of 3 points deducted from the final grade for the course unit.
| Date | Time Slot | Course title | Lecturer | Location |
| 08/10/2026 | 9h00-12h00 | Understanding generative AI | Mohamed Kadri | SophiaTech campus, Lucioles |
| 15/10/2026 | 8h30-13h00 | LLM principles and local models | Nikan Zandian Jazi | SophiaTech campus, Lucioles |
| 22/10/2026 | 8h30-13h00 | RAG | Nikan Zandian Jazi | SophiaTech campus, Lucioles |
| 05/11/2026 | 9h00-12h00 | Introduction to agentic AI | Mohamed Kadri | SophiaTech campus, Lucioles |
| 12/11/2026 | 9h00-12h00 | Agents, tools and MCP | Mohamed Kadri | SophiaTech campus, Lucioles |
| 19/11/2026 | 8h30-12h30 | Agentic coding | Mohamed Kadri | SophiaTech campus, Lucioles |
| 26/11/2026 | 8h30-12h30 | Agentic coding | Mohamed Kadri | SophiaTech campus, Lucioles |
| 03/12/2026 | 8h30-12h30 | Projects presentation | Mohamed Kadri | SophiaTech campus, Lucioles |