AI Fundamentals: the agentic coding course
AI Fundamentals is an online course that teaches developers a tool-agnostic system for working with AI coding agents such as Claude Code, Codex, and Cursor. It covers clear context, reliable instructions, and validation loops. The course is in pre-launch, and you can join the waitlist below.
Course facts
- What it is
- An online course on agentic coding for developers
- Instructor
- Antoine van der Lee
- Format
- Text-based lessons, interactive quizzes, and short video walkthroughs
- Level
- Entry point. Basic programming knowledge helps, no prior AI experience required
- Tools
- Tool-agnostic. Applies to Claude Code, Codex, Cursor, and other agents
- Access
- Lifetime access, including future updates and new content
- Certificate
- Certificate of completion
- Refund
- 14-day money-back guarantee
- Price
- Not announced yet. Waitlist members hear first
- Launch date
- Not announced yet. Waitlist members hear first
What you learn: modules and lessons
Each module focuses on fundamentals that keep applying when tools and models change.
Course introduction
What you can expect from the course, why it exists, and which practical experience shaped it.
- Welcome to AI Fundamentals
- Why I created this course
- My journey to agentic coding
Understanding agentic coding
How you build a clear mental model of agentic coding before touching workflows or tools.
- What is Agentic Coding?
- What are AI Agents?
- Agentic coding vs. autocomplete and AI coding assistants
- Why agent guidelines matter more than code
- How agentic coding changes the way you build software
The agentic coding mindset
How you prepare for the mental shift and risks of agentic coding.
- Preparing yourself for a mental shift
- Code less, build more
- Why agentic coding intimidates senior developers
- Treat your agents like junior hires
- Garbage in, garbage out
- Shipping code you didn’t read
- The risk of rapid tech debt
- Deep work in an agentic workflow
Getting started with agentic coding
How you pick an entry point, choose your tools, and get the sample project running without overthinking.
- The AI development tooling landscape
- How to pick your agent of choice (Codex, Claude, Cursor, etc.)
- CLI vs IDE: Which fits you?
- Local vs cloud agents
- Choosing the right model for the task
- Setting up the sample project
Task context, prompts, and execution
How you use the core mechanics that determine whether agents succeed or fail.
- What is a task context?
- Understanding task context limitations
- Understanding tokens and what they cost
- Prompting techniques that save context and tokens
- What is a prompt (and what is it not)?
- Different modes of execution (plan, execute)
- Plan, review, execute, review, compound
- Recursive work and recursive tasks
Building and improving agent instructions
How you systemize experience into reusable agent behavior.
- Why agent instructions matter
- AGENTS.md
- Agent skills (local and public)
- AGENTS.md vs. agent skills: When to use which
- Using agent skills to apply coding principles
- What are MCPs and when do they matter?
- Adding MCP servers
- Why documenting context matters more than ever
Validation, safety, and feedback loops
How you make agentic coding safe, scalable, and production-ready.
- Preventing tech debt by cleaning up frequently
- Commits as checkpoints
- Independent code reviews after a task
- Using linters and formatters
- Giving AI eyes
- Adding AI validation: builds, warnings, and errors
- Lint, format, test, fix
- Security, privacy, and redaction
- Comparing your result against the final project
Scaling across projects and environments
How you apply agentic coding beyond a single repository or project.
- Creating reusable commands
- Running agents in parallel
- Maintaining best practices across multiple projects
- Cross-repository tasks
- Should you use a monorepo?
- Strategic decisions with agentic coding in mind
- Spotting wasted time and optimizing your workflow
- Preparing prompts with prompts
Wrapping up and certification
How you turn the course into a personal agentic coding system and complete your certification.
- Reflecting on your agentic coding system
- Obtaining your certificate
Who the course is for
This is for you if
- You use AI agents like Claude Code, Codex, or Cursor and want more control over the results.
- You feel overwhelmed by how fast AI moves and want a stable way of working.
- You care about quality, risk, and tech debt, not just speed.
- You want fundamentals that still apply when your tool or model changes.
Probably not for you if
- You want a deep dive into one specific tool’s UI or settings.
- You are looking for prompt tricks or a “10x” shortcut.
Who teaches it
The course is created by Antoine van der Lee, the developer behind RocketSim and RocketTrace and the author of the SwiftLee blog. It comes from how he works with AI agents every day, and he measured what works before teaching it, like in his analysis of 1,674 agent sessions. You can also try the free token cleanup prompt to see the kind of guidance to expect, or read the articles and the glossary.
Frequently asked questions
Is this course suitable for beginners?
Yes. This course is designed as an entry point if you feel overwhelmed by how fast AI moves. You will leave with a clear, repeatable way of working.
What are the prerequisites for this course?
Basic programming knowledge is helpful, but no prior AI experience is required. The course focuses on fundamentals that apply to any development workflow.
How long will I have access to the course materials?
You get lifetime access to all course materials, including future updates and new content.
Can I get a refund if I'm not satisfied?
Yes, we offer a 14-day money-back guarantee. If you're not satisfied, we'll provide a full refund.
Is there a certificate of completion?
Yes, you will receive a certificate of completion after finishing the course.
Will this be outdated when the tools change?
That is the point of the fundamentals. Tools and models change every month, but clear context, reliable instructions, and validation loops keep applying. The course teaches that system instead of one tool's settings.
Why pay for a course when so much about AI is free?
Free tips are scattered and often tied to one tool. The course puts the fundamentals in a clear order, with examples, quizzes, and a repeatable way of working. The free token cleanup prompt shows the kind of practical guidance to expect.
Does it matter which tools (Codex, Cursor, etc.) I use?
No. The course is tool-agnostic and focuses on fundamentals that apply no matter which agent or editor you prefer.