👋 Webinar | Moving on from WinDev: an easier migration thanks to AI · Wednesday 21/10 at 9h15 · Sign up →

[AI Training for Tech Teams]

Train your developers in AI, using their own code

Our trainers are developers and AI Engineers. They build each training session for the team attending it, based on your stack and repositories, then teach your developers to work with a code agent and to build applications and agents on Large Language Models.
Trainers who have been using Large Language Models since their beginnings and code with AI every day
A training session built for your team, based on your repositories and your stack
From the first code agent up to your AI agents in production
Receive a programmeSee the formats
Two smiling developers working together in front of a screen displaying code
They trained their teams with Galadrim
180+engineers and experts at Galadrim
40AI engineers from Polytechnique, Centrale or ENS
150+AI solutions deployed with our clients
[The programme]

Four areas, a programme designed with you

We choose with your Tech Lead the areas to cover and the depth of each, based on the team's level and current projects.
Coding with a code agent
Claude CodeCursorGitHub Copilot

Coding with a code agent

Your developers discover what code agents can do and integrate it into their own way of working. They learn to manage context, to identify tasks that benefit from delegation, and to maintain control over what the agent produces.
Build an application on a LLM
LLM APIsRAGEvaluation

Build an application on a LLM

Your teams learn to identify business problems that an LLM solves well and to integrate them into your applications. Our trainers share what they've learned by designing over 150 AI solutions for a wide variety of businesses.
Develop AI agents
AgentsMCPGo-live

Develop AI agents

Giving tools to a model, chaining several steps, and framing its actions. Your teams build an agent on a specific company case and see what is needed to put it into production.
Discover your AI optimisation levers
Code AuditMCPOrganisation

Discover your AI optimisation levers

At Galadrim, connecting AI to our code and tools has revealed major optimisation levers. We train your CTOs and CIOs to implement the same approach.
[By profile]

What each profile learns to do with AI

Three examples of exercises per profile. We then replace them with tasks drawn from your projects.
01

Developers

Delegating a feature to a code agent
Writing and completing tests
Understanding an unfamiliar codebase
  • Claude Code
  • Cursor
  • GitHub Copilot
02

Tech leads

Writing the team's shared context
Reviewing an AI-produced pull request
Breaking down a project for an agent
  • Claude Code
  • Cursor
  • GitHub
03

Data engineers and data scientists

Writing and optimising SQL queries
Building a RAG on your documents
Evaluating a LLM's responses
  • Python
  • LLM APIs
  • notebooks
04

DevOps Engineers and SREs

Writing CI/CD pipelines
Analysing logs during an incident
Automating operational tasks
  • Claude Code
  • GitHub Actions
  • Terraform
05

QA Engineers

Drafting scenarios from specifications
Automating end-to-end tests
Reproducing and documenting a bug
  • Playwright
  • Cypress
  • GitHub Copilot
06

CTOs and CIOs

Choosing the team's tools and licences
Define code usage rules
Measure the impact on delivery
  • Claude
  • ChatGPT
  • GitHub Copilot
[Our method]

Training tailored to your projects

Three steps, from the initial discussion with your tech team to usage monitoring.

A discussion with your tech team

We review your stack, repositories, authorised tools, and team's proficiency with a Tech Lead. Together, we decide whether the exercises will focus on your code or a repository prepared to mirror your project.

Tailor-made sessions, practice-focused

The trainer builds each session for your team. Everyone practises at their workstation in an environment similar to yours, with the trainer available to help them.

An up-to-date vision of agentic development

Your developers leave with a clear vision of the state of the art in agentic development, as well as with the training materials and exercises. One month later, we follow up with you. A discussion channel remains open in the meantime for their questions.
[Client Cases]

They trained their tech teams with Galadrim

A team of developers and IT management trained in AI agents.
SOFTWARE PUBLISHEROpen Informatique

Train a team of developers in the daily use of a code agent

Open Informatique develops bespoke management software using WinDev, the development environment from French publisher PC Soft. Galadrim has built a three-session training course for its developers to use code agents daily.
On the agenda: how a language model works, the installation and piloting of Claude Code, then writing their own skills, with exercises on code repositories prepared for the team.
10/10of satisfaction
3sessions
11hof training
Open Informatique Logo
FUNERAL SERVICESOGF

Train the 70 employees of an IT department in agentic systems

OGF is the leading French funeral services operator. Galadrim presented agentic systems to the 70 employees of its IT department.
On the agenda: the four main agent architectures, from agentic workflow to MCP protocol, then two projects in production at our clients', a product data sheet generator and an email processing agent. The session concluded with questions on productivity gains and the control of an agent's results.
70IT department employees trained
4agentic architectures explained
2projects in production analysed
OGF Logo
[Formats and Durations]

A format for each objective

  • 1 to 3 days
    Developers and Tech Teams

    Bespoke training

    Exercises built on your stack and repositories, with most of the time spent coding. The training takes place intensively or in four sessions spread over six to eight weeks. After the training, a follow-up accompanies each developer on their own projects.
  • 1 day
    Tech or mixed teams

    AI Hackathon

    Your teams prototype AI applications or agents on topics relevant to your company. One day to create collective momentum and foster new ideas.
  • Ambassador developers

    Training for AI Referents

    Volunteer developers become AI relays within the tech team and support their colleagues daily.
01/03
[By sector]

The same training courses, with your sector's constraints

A tech team from a bank, software company or industry works on very different systems, data, and compliance requirements. We incorporate these constraints into the exercises: the data your developers handle, the rules governing your code, and the systems it interacts with. We also train other roles within your company, with the same logic.
View training courses by sector
DSC7123

Receive a programme tailored to your stack

Receive my programme
[FAQ]

What tech teams ask us

We start from their current uses to go further: entrusting a complete task to an agent, giving it the right context, breaking down the work and efficiently reviewing what it produces. Developers who are already comfortable progress to building agents.
Yes, provided you know how to use it. Our training courses present agent-based development in depth, including limitations and risks. Your developers discover proven methods there to deliver faster and tackle more ambitious topics, while maintaining control of their code.
Familiarity with coding is sufficient for the code agent training. For LLM applications and AI agents, it's beneficial to know a language like Python or TypeScript and to have previously called an API. We calibrate the programme during our initial discussion.
The choice is yours. We work on your repositories using the tools authorised in your company, or on a practice repository that mirrors your tech stack and project structure. We prepare the exercises in your languages and frameworks, including in less common environments, such as WinDev at Open Informatique.
Claude Code, Cursor, GitHub Copilot or Codex for code agents, and Anthropic, OpenAI or Mistral APIs for applications and agents. We prioritise using the tools your team already has. If you're starting out, we'll help you choose licences suitable for professional use.
This is a recurring theme throughout each training programme. Participants learn to have tests written, to review generated code, to protect secrets, and to verify dependencies added by the agent.
Yes. Our engineers develop AI applications and agents for our clients, so a project initiated during the training can become a larger undertaking carried out with your teams.
Depending on your company size and your skills operator, all or part of the training may be covered. We help you prepare the application from the initial scoping.
The price depends on the number of participants, the chosen format, and the preparation time for exercises using your code. Tell us who you wish to train, and you'll receive a detailed quote within 24 working hours.
[Contact us]

Let's discuss training your tech team.

Tell us how many people you wish to train, your stack, and the tools you use. A trainer will call you back to build the programme with you.

Over 800 companies have trusted us to create their web, mobile, and AI products

Your enquiry

We'll get back to you within 1 hour.