Code
MPYF MKT 6443
Level
M2
Field
Marketing, commercial
Language
Français/French
ECTS Credits
2
Class hours
28
Total student load
40
Program Manager(s)
Department
- Management, Marketing et Stratégie
Introduction to the module
This course aims to train students in the dual competency of AI Manager and AI Marketing Engineer. The objective is not merely to skim through existing tools, but to adopt a Spec-Driven Development methodology: designing, structuring, and orchestrating AI products and agents that address specific marketing needs.
The guiding thread: the approach is resolutely centered on the customer and the marketing product. Technique comes into play in a second phase as a facilitator, illustrated through clear architecture diagrams.
Learning goals
- 1. S’approprier les usages avancés et spécialisés des outils de l’intelligence digitale en s’assurant de leur impact durable et responsable
Learning objectives
- 1.1 - Audit advanced and specialised uses of digital intelligence tools in order to deploy them appropriately, taking into account the strategic context of organisations.
- 1.2 - Use digital intelligence tools efficiently to support the societal, digital, energy and environmental transformations of organisations, ensuring their sustainable and responsible impact.
Course Learning objectives
• Maîtriser le prompt engineering avancé
• Sélectionner et orchestrer des LLMs selon les besoins d’un cas d’usage marketing
• Concevoir des architectures hybrides combinant workflows déterministes et agents IA natifs
• Définir et mesurer les KPIs de performance spécifiques aux systèmes IA
• Cadrer un projet IA en intégrant gouvernance, validation des données et conformité
• Master advanced prompt engineering
• Select and orchestrate LLMs based on the needs of a marketing use case
• Design hybrid architectures combining deterministic workflows and native AI agents
• Define and measure performance KPIs specific to AI systems
• Scope an AI project by integrating governance, data validation, and compliance
Content : structure and schedule
1. Advanced & Systemic Prompt Engineering
• Content: Structuring industrial-grade prompts (Context, Role, Objectives, Constraints).
• Efficiency tools: Using and creating prompt generators and evaluators to automate response quality.
2. LLM Orchestration
• Content: Understanding the current large language model ecosystem. How to choose the right LLM for a given use case (cost, latency, creativity, logic).
• Practice: Juggling and coordinating multiple LLMs within a single ecosystem.
3. Automation: Workflow vs Native AI
• Content: Strategic trade-offs. When should deterministic workflows (n8n, Make) be used, and when should autonomy be given to Native AI (Agents)?
• Practice: Hybridizing both approaches to maximize reliability.
4. Performance KPIs & Agent Evaluation
• Content: Evaluating AI differs from evaluating a conventional product. Defining and implementing new performance KPIs specific to agents (hallucination rate, completion accuracy, automation ROI).
5. Scoping, Governance & Data Validation
• Content: Exclusive focus on scoping an AI project. Bias management, securing and validating input/output data, corporate data compliance and governance, illustrated through real-world use cases.
Sustainable Development Goals
Ce cours développe chez les étudiants une double expertise (management de l’IA et ingénierie marketing IA) directement alignée sur l’ODD 4 (Éducation de qualité), en les formant à des compétences professionnelles émergentes et transférables sur le marché du travail. Il contribue également à l’ODD 9 (Industrie, innovation et infrastructure), en les entraînant à concevoir et déployer des agents IA appliqués à des cas d’usage marketing concrets, favorisant ainsi la diffusion de l’innovation technologique au sein des organisations.
This course develops in students a dual expertise (AI management and AI marketing engineering) directly aligned with SDG 4 (Quality Education), by training them in emerging professional skills that are transferable to the job market. It also contributes to SDG 9 (Industry, Innovation and Infrastructure), by training them to design and deploy AI agents applied to concrete marketing use cases, thereby fostering the diffusion of technological innovation within organizations.
Number of SDG's addressed among the 17
2
Learning delivery
synchrone
Pedagogical methods
Les sessions alterneront entre théorie marketing et manipulation directe dans les labs de pointe :
• Environnement Microsoft : Immersion via Microsoft Labs (modules Microsoft Learn IA Product).
• Écosystème Lab & Agents : Exploration et benchmark des solutions clés du marché : Manus.AI, Google (Gemini Antigravity), et Claude (Cowork).
Sessions will alternate between marketing theory and hands-on practice in cutting-edge labs:
• Microsoft Environment: Immersion via Microsoft Labs (Microsoft Learn AI Product modules).
• Lab & Agents Ecosystem: Exploration and benchmarking of key market solutions: Manus.AI, Google (Gemini Antigravity), and Claude (Cowork).
Evaluation and grading system and catch up exams
The evaluation validates both individual technical mastery and the ability to collaborate on a finished product.
• 60%: Individual Grade (Final Exam)
o Format: Theoretical scoping questions and technical tasks to be completed individually (validating a complex prompt, workflow architecture diagram, data governance choices).
• 40%: Group Grade (Group Project)
o Deliverable: Creation and presentation of a conversational agent applied to a concrete marketing use case (hybrid approach).
o Topics: Students will work either on a real company use case or from a list of targeted topics provided on the first day.
The school’s current academic regulations constitute the reference document. If the final grade is below 10 out of 20, a catch-up assessment is organized and counts for 100% of the final grade. This catch-up assessment will consist of an individual research, reflection, and application portfolio, or an oral exam, on a course-related topic. Any assignment submitted after the deadline will be graded 0. Any assignment submitted after the deadline will be graded 0. Grades may be individualized based on participation (unexcused lateness or absence in class, behavior in class, etc.) in the form of a bonus or penalty. Punctuality and attendance are an integral part of the shared living and learning environment; they reflect respect for the work of instructors, the group, and the school’s requirements, as well as the professional attitude expected of students. Any unjustified half-day absence will result in a one-point penalty on the final grade for the module concerned. This penalty may also apply to any unjustified lateness. In cases of repeated lateness or absences, the number of points deducted may be increased to ensure the smooth running of classes.
Module Policies
The school’s current academic regulations constitute the reference document.
Instructor-Learner Communication
● The instructor will contact learners via their IMT-BS/TSP school email address and the Moodle portal. No communication will take place via personal email addresses. It is the student’s responsibility to check their IMT-BS/TSP mailbox regularly.
● Learners may contact the instructor by emailing their institutional address. If needed, it is possible to meet the instructor in their office or via video conference by appointment.
Learners with Accommodation Needs
If a learner has a disability that prevents them from completing the work described, or that requires any type of accommodation, it is their responsibility to inform the Director of Studies (with supporting documentation) as soon as possible. Learners should also feel free to discuss this with their instructor.
Classroom Behavior
● As a matter of courtesy to the instructor and other learners, all mobile phones, electronic games, or other sound-generating devices must be turned off during class.
● Learners must avoid any disruptive or disrespectful behavior, such as: arriving late to class, leaving early, thoughtless conduct (e.g., sleeping, reading material unrelated to the course, using vulgar language, talking excessively, eating, drinking, etc.). A warning may be given for a first violation of these rules. Offenders will be penalized and may be expelled from class and/or subject to other disciplinary proceedings.
● No lateness is tolerated. Attendance will be recorded on Moodle via a QR code provided by the instructor at the start of each class.
● Learners must arrive on time for exams and other assessments. No one will be allowed to enter the classroom once the first person has finished the exam and left the room. There are absolutely no exceptions to this rule. No learner may continue taking an exam once time has expired. No learner may leave the room during an exam unless they have finished and submitted all documents.
● In the case of remote classes, learners must keep their camera on unless otherwise instructed by the teacher.
Code of Ethics
IMT-BS is committed to a policy of academic integrity. Any conduct that compromises this policy may result in academic and/or disciplinary sanctions. Learners must refrain from cheating, lying, plagiarizing, and stealing. This means producing original work and acknowledging any other person whose ideas and printed materials (including those from the Internet) are paraphrased or directly quoted. Any learner who violates or helps another student violate academic conduct standards will be sanctioned in accordance with IMT-BS rules.
Textbook Required and Suggested Readings
Goglin, C., & Mayol, S. (Dirs.). (2025). Le marketing à l’ère des IA génératives: Enjeux et perspectives. Caen, France: EMS Éditions.
Lendi, S. (2026). Le marketing digital à l’ère de l’IA: Contenu, trafic, conversion: Des fondamentaux jusqu’aux dernières avancées des agents IA (2e éd.). Paris, France: Vuibert.
Keywords
Intelligence artificielle générative, agents conversationnels/agents IA, prompt engineering, Spec-Driven Development, architecture de workflow IA, gouvernance des données, AI marketing engineering / Generative artificial intelligence, conversational agents/AI agents, prompt engineering, Spec-Driven Development, AI workflow architecture, data governance, AI marketing engineering.