Training

Training

Three courses for technical teams: Claude Code (2 days), information systems architecture (5 days) and local AI (2 days). In-house, on site or remote.

Courses designed by a practitioner, for teams who want to apply what they learn the very next day. Each session mixes input, demonstrations and hands-on work on cases close to your context.

Building software with Claude Code

Duration: 2 days · Developers, tech leads, architects

Claude Code is a coding agent that works directly in the terminal and the code repository. Used well, it changes how code is designed, written and reviewed. Used without a framework, it quickly produces a lot of hard-to-maintain code. This course teaches how to get the best out of it, as a team and safely.

Objectives

  • Integrate Claude Code into a development team’s daily work;
  • Give the agent the project context and clear rules;
  • Connect it to company tools and automate repetitive tasks;
  • Stay in control: review, testing, security and code confidentiality.

Programme

  • Day 1 — Principles of agent-assisted development; installation and first uses; the CLAUDE.md file and project conventions; permission management; from requirement to pull request: explore, plan, code, test.
  • Day 2 — Custom commands and skills; connecting tools through MCP; hooks and subagents; CI integration; review, security and governance best practices; workshop on the team’s own repository.

Information systems architecture

Duration: 5 days · Architects, tech leads, domain leads, future architects

A complete course to design, document and evolve an information system, drawing on thirty years of practice, from large enterprises to SaaS vendors.

Objectives

  • Master the architecture layers: business, functional, application, technical;
  • Map an existing information system and build a transformation roadmap;
  • Choose and justify architecture styles: microservices, event-driven, APIs;
  • Set up light, effective architecture governance.

Programme

  • Day 1 — The architect’s role; architecture layers and views; overview of reference frameworks (TOGAF).
  • Day 2 — Mapping and planning the IT landscape; analysing the existing estate; technical debt.
  • Day 3 — Architecture styles: modular monolith, microservices, event-driven architectures, API management.
  • Day 4 — Cloud, platforms and non-functional requirements: security, performance, resilience, costs, operations.
  • Day 5 — Deciding and documenting: architecture decision records (ADRs), architecture documents, boards; integrating AI into the information system; end-to-end case study.

Local AI

Duration: 2 days · Architects, developers, infrastructure and security teams

Running AI models on your own infrastructure keeps your data and costs under control. You still need to know how to choose models, hardware and architecture. This course gives you the keys to decide, then to deploy.

Objectives

  • Know when local AI makes sense, and when it does not;
  • Choose models, hardware and inference engines to fit your needs;
  • Deploy production-ready local AI: API, RAG, security, monitoring;
  • Compare quality and costs with managed services.

Programme

  • Day 1 — Sovereignty and confidentiality stakes; overview of open-weight models; hardware sizing (GPU, memory, quantisation); inference engines; first deployments.
  • Day 2 — Exposing a model through an API; RAG over internal documents; agents and tools; security, monitoring and evaluation; cost and quality comparison with managed services; architecture workshop.

Practicalities

  • Format: in-house, on site or remote, 4 to 10 participants.
  • Tailoring: content is adapted to your context, tools and team level after an initial conversation.
  • Price: on quotation, depending on format and number of participants.

Book a scoping call