COURSES
AI and the Legal Profession: Practical Applications and Requirements of the AI Act
Overview of AI use in legal practice, the limits of generative AI and the need for human review, confidentiality and data sovereignty, AI Act obligations for legal professionals, and how to build an AI literacy plan, closing with a workshop combining a practical exercise and a risk classification task.
- Identify relevant AI uses in a legal context
- Recognise the limits and risks of generative AI applied to law
- Understand the AI Act obligations that apply to legal professionals, including Article 4
- Adopt a systematic verification approach toward AI-generated output
- Build an AI literacy plan suited to a legal team
The AI Act Applied to Engineering, Construction, and Real Estate
Overview of the AI Act framework, use cases in engineering, construction and real estate, risk classification of AI systems, an update on the Digital Omnibus and how to build an AI literacy plan, closing with a hands-on workshop mapping an organisation’s AI systems.
- Understand the Article 4 AI literacy obligation as it applies to engineering, construction and real estate
- Identify and classify the AI systems used in these fields according to the AI Act risk level
- Distinguish AI uses that create immediate value from those that call for careful oversight
- Build an AI literacy plan suited to a technical team
- Track recent regulatory developments and their impact on the compliance timeline
Document Governance for AI Systems
Why AI systems need documentation, the AI Act’s documentation requirements by risk level, how to structure a technical documentation sheet, building an AI usage register, and sharing documentation roles within an SME, closing with a hands-on workshop drafting a documentation sheet and a register entry.
- Understand the AI Act’s documentation requirements according to a system’s risk level
- Draft a clear and complete technical documentation sheet (model card)
- Build an AI usage register within an organisation
- Share out documentation responsibilities without creating a disproportionate burden for an SME
Building Production AI Agents with LangGraph
Overview of agent frameworks and how to choose between LangGraph and CrewAI, agent architecture (state, memory, tools), building an agent with LangGraph, integrating external APIs, traceability, human checkpoints and error handling, closing with a hands-on workshop building a complete multi-step agent.
- Understand the architectural differences between LangGraph and CrewAI and choose the right tool for the context
- Build a structured AI agent with state and tool management
- Integrate human checkpoints for sensitive decisions
- Set up traceability sufficient for real-world use
- Identify what it takes to move beyond a demo
RAG in production
A quick refresher on RAG foundations, what separates a prototype from a production system, evaluating answer quality, managing the document lifecycle, cost and performance trade-offs, and data sovereignty, closing with a hands-on workshop industrialising an existing RAG pipeline.
- Distinguish the requirements of a RAG prototype from those of a production system
- Evaluate the quality and reliability of a RAG system’s answers
- Manage the lifecycle of source documents in a RAG pipeline
- Weigh up cost, performance and data sovereignty
- Industrialise an existing RAG pipeline with quality tests
AI and Project Deviation Detection
Why drift is hard to catch early, which data sources to use, basic analysis concepts (time series, anomaly detection), building early warning indicators, and how to present results without judging existing methods, closing with a hands-on workshop building a drift indicator in a spreadsheet and then in Python.
- Understand which data sources can be used to detect schedule and budget drift
- Build early warning indicators from past project data
- Distinguish situations where a simple statistical method is enough from those that call for a more complex approach
- Present a detection tool’s results as decision support, without judging existing methods
- Build a first drift detection indicator on a concrete case
AI for Finance and Insurance
AI adoption in the Luxembourg financial sector according to the CSSF and BCL review, regulated use cases (KYC, credit scoring, fraud and money laundering detection), the distinction between perceived and actual risk under the AI Act, the landscape of competent authorities (CSSF, CAA, CNPD), and common governance gaps, closing with a hands-on workshop assessing a fictional use case.
- Understand the state of AI adoption in the Luxembourg financial sector according to the CSSF/BCL thematic review
- Distinguish use cases genuinely classified as high-risk under the AI Act from those merely perceived as such
- Identify the role of each competent authority (CSSF, CAA, CNPD) depending on the type of institution
- Spot the most common governance gaps in regulated AI systems
- Assess a regulated AI use case against the applicable requirements
Customized Training
Whatever the needs may be, we are able to develop a customized program for decision-makers, executives, and employees, taking into account their objectives and constraints.
