Developer training
Claude Code training for development teams
A coding agent reads the repository, runs commands and edits several files. The decisive skill is no longer writing every line but scoping, testing, reviewing and securing. In two days, your team learns to do it on its own code.
- Duration
- 2 days (14 hours)
- Level
- Intermediate
- Format
- On-site / Remote
- Price excl. VAT
- €3,000 excl. VAT in-house, per group€1,000 excl. VAT open session, per participant
Why this training
Code generated faster is not better software
Trying the tool alone teaches the tool; it does not teach teamwork. A team adopting an agent on a shared repository must decide who writes the repository instructions, which commands the agent runs on its own, who reviews, and how you know a task is done.
We speak as users. In our AI-First Delivery offer, AI generates the code; our engineers and AI PhDs review and validate it, to put custom software into production in 4 weeks. These are the practices the training passes on.
From vibe coding to maintained code
Accepting generated code without reading it is fine for a throwaway prototype, not for software the company is accountable for. The training teaches you to keep control of the code.
A passing test is not proof of security
According to Veracode (March 2026), generated code is syntactically correct in over 95% of cases, yet a known vulnerability is introduced in 45% of tasks when no security instruction is given.
Gains are measured, not felt
In the randomised trial METR published in July 2025, experienced developers took 19% longer with AI, while believing afterwards that they had been 20% faster.
Audience and prerequisites
For developers who already know how to code
An intermediate-level course: it does not reteach programming, it teaches how to supervise a coding agent.
- Developers, lead developers and technical managers
- Teams already using a coding assistant or agent, each with their own settings
- Teams preparing to adopt a coding agent on a shared repository
Coding with AI
- Prerequisites
- Software development experience
- Level
- Intermediate
Objectives
What the team can do after the two days
- 01Choose between autocompletion and a coding agent depending on the task
- 02Write and maintain the repository context file (CLAUDE.md, AGENTS.md)
- 03Break work into tasks an agent can execute, with a verifiable success criterion
- 04Review a diff produced by an agent and have the failing test written before the fix
- 05Configure permissions and secrets, and check dependencies added by the agent
- 06Track costs and measure the real gain before and after adoption
Programme
Five modules, on your repository
Two days (14 hours), on-site or remote. In-house, exercises use the team's code and the tool it has chosen.
Module 1
What AI changes in development: from autocompletion to coding agents
What changes when the tool reads the repository, runs commands and edits several files; how the developer's role shifts; setting up on the team's repository.
Module 2
Assisted code generation: scope, plan, test first
The repository context file (CLAUDE.md, AGENTS.md), breaking work into executable tasks, the explore, plan, code sequence; having the failing test written before the fix.
Module 3
Coding assistants and agents: Claude Code, OpenCode, GitHub Copilot
Strengths, limits and selection criteria: supported models, code confidentiality, the option of a local or open-weight model, integration with the editor and the CI pipeline.
Module 4
Code review, debugging and refactoring
Reviewing a diff produced by an agent, having it reviewed by a session that did not write the code, requiring human review before merge; debugging and refactoring under tests.
Module 5
Security and good practice
Permission modes, secrets out of the agent's reach, sandboxing, prompt injection, invented dependencies; cost tracking and gain indicators measured before and after.
Practical details
Duration, format and prices
- Duration
- 2 days (14 hours)
- Level
- Intermediate
- Format
- On-site / Remote
- In-house
- €3,000 excl. VATfor a group from the same company
- Open session
- €1,000 excl. VATper participant, inter-company session
In-house, the training runs on the team's repository and the tool it has chosen: it leaves behind a context file, a shared configuration and a review rule.
All prices are exclusive of VAT.
Funding
Getting the training funded
JustAI's Qualiopi certification gives access to funding from your French skills operator (OPCO). We provide the quote, the detailed programme and the training agreement: the documents your OPCO expects.
OPCO
Your skills operator (Atlas, Opco 2i, Akto, Afdas…) can cover all or part of the training, in-house or in an open session, under the skills development plan. Each OPCO applies its own rules and ceilings.
Payment: what changes on 1 October 2026
At several OPCOs, including Atlas, AKTO and Opco EP, direct payment of the training provider becomes the exception: the company pays the invoice, then claims reimbursement. Check your OPCO's rule before committing to the training.
CPF
This training is not certified, so it is not eligible for the CPF (France's personal training account). The CPF can fund certified training, such as our 4-day programme preparing for the RS No. 6891 certification on producing and reviewing professional content with generative AI, intended for content-production roles and subject to the applicable cap.
We help you put the funding application together: it is part of the service.
OPCO: what changes on 1 October 2026 (in French)Quality
A certified training provider, an engineering firm
Qualiopi Certification
Certified for professional training actions. Certificate valid from 08/07/2024 to 08/06/2027.
View the certificateWe practise what we teach
JustAI is an AI consulting and engineering firm, based in Paris since 2020. We run our own GPU servers in France, serve open-weight models through our own inference APIs and ship custom software to production. Our training draws on that practice.
Our infrastructureFAQ
Your questions about this training
Is the training only about Claude Code?
No. The foundation (scoping, tests, review, security) is common to coding agents. We compare Claude Code, OpenCode and GitHub Copilot; in-house, exercises use the tool your team has chosen and your own repository.
Is this a vibe coding course?
Not in the usual sense. Accepting generated code without reading it suits a throwaway prototype, not software the company is accountable for. The training teaches you to benefit from generation speed without losing control of the code.
Can our code stay confidential?
It is one of the tool selection criteria covered in the training. Claude Code works with Anthropic's Claude models; OpenCode can run on a locally hosted model. For sensitive projects, JustAI runs its own GPU servers in France and its own inference APIs on open-weight models.
What level is required?
Software development experience. The course is intermediate: it is for developers who can code and want to work with an agent without losing code quality.
How much does it cost and how can it be funded?
Two days (14 hours): €3,000 excl. VAT in-house for a group from the same company, €1,000 excl. VAT per participant in an open session. It can be funded by your OPCO under the skills development plan; it is not eligible for the CPF.
Related training
Build the pathway
3 days (21 hours) · Advanced / 4 days (28 hours) · Expert
AI agents training
See the training1 day (7 hours) · Advanced
Prompt engineering training
See the training3 days (21 hours) · Advanced
The technical stakes of AI for IT departments
See the trainingFurther reading
Related articles
Our articles are published in French.
Outils & technologies
Claude Code en équipe : ce que vos développeurs doivent apprendre
Read the articleAnalyse
Claude Code vs Hermes vs Opencode : lequel choisir en 2026 ?
Read the articleAnalyse
Xirp : Spotify ouvre son orchestrateur d'agents de code (Claude Code, Gemini CLI, Codex)
Read the articleOutils & technologies