Beginner tier · no experience needed
Twenty lessons take you from “what is a computer?” to job-ready full-stack developer: HTML to React, Node.js to PostgreSQL, Docker to production. AI is built into every chapter as a teammate you learn to check, never an autopilot you trust blindly.
✓ You already own this bootcamp. It’s waiting in your Active Bootcamps.
✓ Build succeeded · prior experience required: none
the job
They build every layer of a web product: the screens people click, the server logic behind them, the data underneath and the cloud it all runs on. The “AI-powered” part means two things. You use AI coding agents to work faster, and you build AI features into the product itself. Follow one click through Meridian, the app you’ll build, to see every layer you’ll learn.
Dana clicks “Mark as paid” in Meridian.
Underneath every layer: Git (L6), architecture (L9), debugging (L13), testing (L14) and reading code you didn’t write (L15).
A typical week on the job
Turn a client request into an issue, a plan and an estimate
L6 · L15Build the feature across the UI, the API and the database
L8 · L10 · L11Review an AI-generated pull request and catch what it missed
L6 · L16Trace a production bug from the logs to its root cause
L13 · L19Ship behind a feature flag and watch the dashboards
L19is this you?
This is a Beginner-tier bootcamp. It assumes no IT background at all, and it’s honest about who it isn’t for. Here’s the diff.
You like building things people can see and click, and working out why they break
You enjoy both sides of software: how it looks and how it works
You’re curious about AI and want to use it well, not just use it a lot
You’re comfortable saying “I don’t know yet”, and then finding out
You’re starting from zero, or switching careers from a non-IT field
You’d rather read, build and repeat at your own pace than follow a cohort timetable
You want a portfolio of shipped projects, not a certificate of attendance
You want material you can come back to in your first developer job
You need live classes or an instructor to keep you accountable. There are none.
You want video lessons. Everything is written, like real documentation.
You already ship production code. The early lessons will feel slow.
You want a quick overview. This is 341 chapters of real depth.
You hope AI will do the learning for you. The method makes you solve it first.
the continuous product
In Lesson 1 you meet Dana Okafor, founder of Northlight Consulting. Her remote-first team of about twelve tracks clients, projects, tasks and invoices across spreadsheets, email threads and sticky notes. Your job is to build Meridian, the one app that replaces all of it. Every lesson ends with a new Meridian chapter, from the first brief to production launch and handover. Drag the slider to watch it grow.
Dana Okafor, founder of Northlight Consulting, explains the mess: clients, projects, tasks and invoices scattered across spreadsheets, email threads and sticky notes.
the curriculum
Every lesson has 16 to 22 chapters and ends the same way: a Broken Build Clinic to repair, then the next Meridian release. Open any lesson to see every chapter inside it.
inside every chapter
Five learning sections build the concept. Then come a senior developer’s commentary, an AI-collaboration exercise, a realistic practice task and a 30-question assessment with an explained answer key. Here’s a real one: Chapter 12.04, Tokens and JWT.
Starts from what you already know, sessions the server has to remember, and introduces the opposite idea: a token that carries everything needed to trust it.
ai, done right
Every chapter’s W section runs the same six-step workflow. You always do the thinking first. Then you use AI to check your work, and you check the AI right back.
## 5.11 async/await in practice - ignored the rejected promise → wrapped in try/catch ## 11.07 Indexes and query performance - suggested an index the query never uses ## 12.04 Tokens and JWT - kept the refresh token in localStorage → XSS risk
Send a clear, specific prompt and read the whole answer. Don’t correct it and don’t copy it. Not yet.
broken build clinic
Most real developer work starts in code someone else wrote. So every lesson has a Broken Build Clinic, from an inherited laptop in Lesson 1 to a misbehaving RAG system in Lesson 18.
An internal shift-scheduling tool for a small logistics company. Priya Anand built it alone over about a year, then left without writing anything down, and Jordan Reyes has kept it alive since. Halcyon turns up throughout the bootcamp. In Lesson 15, when Jordan steps away, it becomes entirely yours.
what you walk away with
One language family end to end, the tools professional teams actually use, and a project in nearly every lesson.
The client product you build across all 20 lessons, from sticky-note brief to v13.
A web app that lets a team ask plain-English questions about their own documents.
One complete backend for a client: routing, validation, logging, background jobs and API docs.
A complete React frontend for a blog platform, then its data layer.
One complete authentication system for a client: password hashing, sessions and role-based access.
Your own AI-powered product, planned, built, tested, deployed and presented.
career launch
The final lesson is a guided capstone plus a complete job-search playbook, from choosing a project that sells you to your first 90 days in the role.
ROLES THIS PREPARES YOU TO APPLY FOR
No bootcamp can promise you a job. This one makes sure you can show your work, explain your decisions and handle the interview.
where it leads
Gross annual base salary by experience level. The ranges are indicative and drawn from public salary data. Your offer will depend on city, company and negotiation.
Your first role: shipping features inside an existing codebase, with code review from seniors. Many junior offers land between €45k and €55k.
You own features end to end, from database to UI. Munich, Frankfurt, Hamburg and Stuttgart typically pay 10–20% above the national average.
Architecture decisions, mentoring and production ownership. Staff and lead roles reach roughly €90k to €120k.
Sources: jobrise.io 2026 full-stack salary guide, Indeed Germany, PayScale Germany (January 2026, 162 profiles).
Amsterdam, Rotterdam, Utrecht and Eindhoven are the main hubs, and English-speaking teams are common.
Lean Dutch product teams value developers who can work across the whole stack instead of one layer of it.
Senior and lead roles at scale-ups and international companies reach the top of this band. Contract rates run higher still.
Sources: Indeed Netherlands (September 2026, 3.3k developer salaries), PayScale Netherlands. Dutch offers are often quoted per month and exclude the statutory 8% holiday allowance.
Brussels, Antwerp and Ghent are the main hubs. Packages often add meal vouchers, holiday pay and a 13th month on top of base salary.
Many experienced developers move to freelance contracts, so employers compete on benefits as much as on base salary.
Senior and lead roles in fintech, logistics and EU-adjacent organisations sit at the top of the band. A company car is common at this level.
Sources: Indeed Belgium (September 2026), PayScale Belgium (July 2026, 28 profiles). Belgium has the highest tax burden of the three, so the same gross salary means less take-home pay.
never stuck, never alone
Self-paced doesn’t mean unsupported. Mentor Bob is an AI study assistant built into every lesson page. It has already read the section you’re on, so you can ask for a simpler explanation or a fresh example the moment you’re stuck.
questions
Yes. That’s exactly who it’s written for. Lesson 1 starts with what a computer actually is, then covers files and folders, the command line, your code editor and how the internet works. You build your first web page, end to end, before Lesson 1 is over.
One language family, end to end: JavaScript and TypeScript. Around it you learn HTML, CSS and Tailwind; React and Next.js; Node.js and Express; SQL with PostgreSQL and Prisma, plus MongoDB; Git, GitHub and GitHub Actions; Playwright for testing; Docker and CI/CD for production; and LLM APIs, embeddings and RAG for AI features.
Not to get started. The first AI exercises run on the free tier of whichever assistant you set up in Lesson 1. Later chapters explain how AI tool pricing and usage costs work, so a paid plan is your call, made with real information.
Because someone still has to decide what to build, check that the code is correct and secure, and keep it running. That judgement is what employers hire for, and every chapter trains it. There are whole chapters on where AI fails (16.12), when not to use it (17.13) and staying a real engineer (16.14).
It’s self-paced, with no deadlines or cohort dates. With 341 chapters, plan for months of steady work rather than a few weeks. Access never expires, so you can pause whenever life gets busy.
No. Every section is written: structured, searchable and easy to come back to on the job, the way real documentation works.
A reasonably modern Windows, macOS or Linux computer. Lesson 1 walks you through installing the full toolchain on each of them.
Ask Mentor Bob, the AI study assistant on every section. It has already read the section you’re on. Each chapter’s practice task also spells out what “done” actually looks like, so you can check your own work.
LLM Engineer for Beginners specialises in LLM systems: fine-tuning, evaluation, agents and deployment. This bootcamp makes you a complete web developer first, across frontend, backend, data and production, with AI as both your coding partner and a feature you ship.
No. You pay €349 once and keep access for life: every lesson, chapter and quiz, plus Mentor Bob.
ready when you are
One-time payment · lifetime access
Secure checkout by Stripe. No subscription and no renewals.