Claude Code training, 1-on-1.You ship software with an AI agent.
A Claude Code expert opens your repo with you and walks through what matters: setup and first agentic session, real dev tasks (features, refactors, tests, bugs), MCP servers, hooks, and automation all the way to CI. We start from your real code, not theory.
★★★★★ 4.7/5 · 300+ pros trained · France Num certified
ActiveCampaign
Adalo
AdCreative.ai
Ahref
Airtable
Allo (The Mobile First Company)
Anthropic
Apify
Apollo.io
AttioAttio Implementation Partner
Base44
Baserow
Brevo
Bright Data
Browse AI
Bubble
CaptainData
ChatGPT
Claude
Claude CodeClaude CoworkClaude Design
Clay
Clickup
Cursor
DeepSeek
Dust
ElevenLabs
Fillout
Flutterflow
Folk CRM
Folk Implementation Partner
Freepik Spaces
Gamma
GeminiWe ship with Claude Code for clients, not just in theory.
Most Claude Code trainings are recorded demos by people who opened the tool the night before. At Hack'celeration it's the opposite: driving the agent on real features, plugging in MCP servers, writing hooks, automating reviews in CI, that's our daily agency work. Everything we teach you, we practice on shipped projects. We know the traps (the agent drifting on a big refactor, the over-broad permissions) because we've already solved them.
- We ship with Claude Code for clients every week, not just in theory
- 1-on-1 format: the trainer adapts to your level, junior dev or seasoned lead
- We tell you when Claude Code isn't the right call (sometimes IDE autocomplete is enough)
- We start from your real repo and your actual tickets, not a toy project
Four pillars so you ship software with an AI agent.
A badly scoped Claude Code means an agent that drifts, diffs applied without review, and an API bill that runs away. Most of the trouble comes from the workflow, not the tool. We pick up your real repo and work through the four pillars together.
- Setup and first session
Claude Code installed, your first agentic session
We install Claude Code in your terminal or IDE (VS Code, JetBrains), connect your account or API key, and set up your project's CLAUDE.md so the agent gets the right context. You learn to start a session, let Claude read your repo, plan, and propose a diff you approve. In one session, you go from zero to an agent running on your real code.
Run my first session - Real dev tasks
Features, refactors, tests, and bug fixes
We drive Claude Code on your real tickets: add a feature, refactor a module, write the missing tests, fix a bug from a stack trace. You learn to scope a task, iterate on the plan, review the diff before applying it, and stay in control. The agent codes fast, but you approve every change.
See the dev tasks - MCP, hooks and commands
MCP servers, hooks and custom commands
We plug in MCP servers to give Claude access to your tools (GitHub, a database, internal docs, a browser). We set up hooks to trigger lint, tests or formatting at each step, and we build your own slash commands for recurring tasks. You turn Claude Code into an agent that actually knows your environment.
See MCP and hooks - Automate and ship
CI, code reviews and automated repo workflows
We take Claude Code from the IDE to the pipeline: automated PR reviews, test generation in CI, GitHub issue resolution, doc updates. We set the guardrails (permissions, human approvals, limits) so you ship without risk. You leave with a clear workflow to automate what repeats, without letting the agent run loose in production.
Talk to a trainer
Meet our trainers, leave with a plan.
Drop your email. We get back to you to connect you with a Hack'celeration-certified trainer: we look at your repo and workflow, spot where Claude Code wins you time, and tell you where to start. No commitment, even if you don't take the training.
- A diagnosis of your stack and dev workflow
- The first tasks to hand to the agent, in priority order
- The right 1-on-1 format for your level and goals
- An honest take: Claude Code or another tool for your case
Your Claude Code program, step by step.
Five steps, no skipping. Each one on your real repo, with a clear deliverable. From the first session we install the agent and scope where it wins you time. By the end, you drive Claude Code on your projects without us.
- Step 1 · Audit of your stack and workflow
We open your real repo and spot where Claude Code wins you time
First session, we share your screen and look at your project together. What language, what framework, what tests, what CI? Where do you spend the most time: boilerplate, debugging, refactors, docs? We install Claude Code in your terminal or IDE, write a first CLAUDE.md tailored to your repo, and list the tasks where the agent will genuinely win you time. You leave with a clean setup and a plan, in priority order. No theory, your real code.
- Step 2 · Your first guided build
We drive Claude Code on a real task, together
We take on a real ticket: a feature, a refactor, or a bug fix. I show you how to scope the request, let Claude plan, iterate on the plan before it touches code, then review the diff line by line before applying it. We also cover how to take back control when the agent drifts, and how to keep a clean git history. By the end, you've shipped a real change with the agent and you know how to stay in control.
- Step 3 · MCP and customization
We plug in MCP servers and tailor Claude to your environment
We give Claude Code access to your world. We install MCP servers for GitHub, your database, internal docs or a browser, set the permissions, and test on your real tools. Then we customize: hooks to run lint and tests automatically, custom slash commands for recurring tasks, tuning the CLAUDE.md. You practice on your own stack, not a dummy example. You finish with an agent that knows your project.
- Step 4 · Automation and CI
We automate your reviews and repo workflows
This is where Claude Code becomes a real team lever. We take it from the IDE to the pipeline: automated PR reviews, test generation in CI, GitHub issue resolution, doc updates when the code changes. We set the guardrails (permissions, human approvals, cost limits) so the agent works without risk. We build the loop: it proposes, you approve, it ships. Control stays with you, not the agent.
- Step 5 · Autonomy
You drive Claude Code on your projects without us
The number one goal: you become autonomous. By the end of the program, you know how to install Claude Code, scope a task, review a diff, plug in MCP servers, write a hook and automate a review. You no longer need an agency to code with an AI agent. And if you want to delegate later, we also run a Claude Code agency, but that's not the point here.
Why train 1-on-1 with us.
- 300+Pros already trained on AI and dev
More than 300 people have gone through our trainings across France and Europe. Devs, tech teams, startup founders, freelancers. Not vanity numbers: people who now ship features with an AI agent instead of watching from the sidelines.
- 4.7/5Rating across 334 verified reviews
Average rating of 4.7 out of 5, across 334 reviews. We won't promise the agent codes everything on its own: you have to scope it, review the diffs and stay in control. But the 1-on-1 format makes the difference in moving from random prompts to a real agentic workflow.
- 1:1A dedicated expert, not a class of 100
You're not a number in a webinar. A trainer opens Claude Code on your real repo, looks at your stack, and works through your actual cases. We schedule sessions around your availability, replays included.
A working agency, recognized by the French State.
Hack'celeration is certified Activateur France Numérique and holds the AI Ambassador title, both granted by France Num to organizations that genuinely drive the digital transformation of companies. On the ground, we ship with Claude Code for clients every week: more than 300 pros trained and a 4.7/5 rating across 334 verified reviews, left by the people who took our programs, not just by the buyer.
- Certified Activateur France Numérique
- AI Ambassador (France Num)
- 300+ pros trained across France and Europe
- 4.7/5 across 334 verified reviews
The questions we get the most.
What is a 1-on-1 Claude Code training?
An individual program with a Claude Code expert, not a class of 100 people. We open the agent on your real repo, look at your stack, and work through your actual cases: setup, real dev tasks, MCP servers, hooks, automation. You ask your questions live, the trainer adapts the pace to your level. We schedule sessions around your availability, and you leave with concrete actions every time. That's what makes the difference between watching a demo and actually shipping with an AI agent.How much does the Claude Code training cost?
There is no single price. We connect you with a trainer certified by Hack'celeration, matched to your need and your level. It varies from one trainer to another, based on their profile and the format that fits your project.Claude Code or Cursor: which one to pick?
Both are good, but not for the same use. Cursor is a full IDE with autocomplete and built-in chat, very smooth for editing code continuously under your eyes. Claude Code is an agent that lives in your terminal (or IDE): it reads your repo, plans a multi-step task, runs commands, and proposes a diff you approve. For agentic building and automation (CI, MCP, hooks), Claude Code goes further. Many devs use both. During the training, we help you decide based on your real workflow.Do I need to know how to code to take the Claude Code training?
Yes, a minimum. Claude Code is a dev tool: it helps you write, refactor and debug code, but you review the diffs, validate the architecture and stay in control. You don't need to be senior, a junior to intermediate dev level is enough. The 1-on-1 format starts from your exact level: if you're starting out, we go step by step; if you're seasoned, we jump straight to MCP, hooks and automation. If you don't code at all, this isn't the right training, we have other no-code tracks.What is an MCP server and why do I need one?
MCP (Model Context Protocol) is an open standard that lets you connect Claude Code to external tools: GitHub, a database, your internal docs, a browser, an API. Without MCP, the agent only sees your local code. With it, it can read a GitHub issue, query your database, or search your docs to do a better job. In the training, we install the MCP servers useful for your project together, set the permissions, and test on your real tools. It's one of the big levers to make the agent genuinely useful.Does Claude Code run in the terminal or the IDE?
Both. At its core, Claude Code is a command-line tool that runs in your terminal, where it has access to git, your scripts and your commands. It also integrates with VS Code and JetBrains IDEs through an extension, to show diffs right in the editor. You can pick based on your comfort: pure terminal to drive by command, IDE to stay in your usual environment. In 1-on-1, we set up the one you prefer and show you how to switch between them.How much does Claude Code API usage cost?
It depends on your access mode. You can use Claude Code through a Claude subscription (Pro or Max), which includes a certain amount of usage, or through the Anthropic API billed per token consumed. The real cost depends on your repo size, session length and the models used. In practice, a well-scoped session costs little, but an agent left without limits can burn through tokens fast. In the training, we show you how to monitor your usage, limit context and set guardrails to keep the budget under control.Can Claude Code break my code or push to production on its own?
Not if you scope it well, and that's the whole point. Claude Code proposes diffs you approve before they apply, and you can require confirmation before each sensitive command. The real risk is giving it too many permissions or approving without reviewing. In the training, we set the guardrails: per-command permissions, human approvals, never an automatic push to production without review. You learn to work with git so you can roll back at any time. Control stays with you.Which languages and frameworks does Claude Code support?
All the common languages: JavaScript and TypeScript, Python, Go, Rust, Java, PHP, Ruby, C#, and most web frameworks (React, Next.js, Vue, Django, Rails, Laravel...). Since the agent reads your repo and config files, it adapts to your existing stack rather than imposing a framework. It's also comfortable with infra (Docker, shell scripts, CI). In the training, we start from your real language and framework, so you work directly in your environment, not a generic example.Is the training online or in person?
100% online, over video, 1-on-1. You join the sessions from anywhere, we share your screen and drive Claude Code on your real repo live. Sessions are recorded if you want to revisit them. The individual format means real interaction: you're not a number in a webinar of 100, the trainer answers your questions about your stack and your tickets. That's what makes learning concrete on a tool you mostly learn by running it on real code.
Your code deserves an agent. Meet your trainer.
Drop your email. An expert who ships with Claude Code daily looks at your repo and shows you how to drive an AI agent on your real tasks. No commitment, even if you don't take the training.