AI Coding for Real Engineers

2-week learning cohort • Starting June 1 • $696

🎉 30% off - Sale ends May 26! Only 3 days left.

It's hard to communicate how much programming has changed due to AI in the last 2 months: not gradually and over time in the 'progress as usual' way, but specifically this last December.

Andrej Karpathy

Do you use AI coding tools?

I do.

Using AI to write software every single day for months has taught me a lot. It's changed some of my opinions, and significantly upgraded the way I think about modern software engineering.

Here's two things I've learned:

1. AI can be both overhyped and powerful. It's extraordinarily useful, when handled correctly.
2. AI is a call for devs like us to deepen & upgrade our fundamental skillset: engineering.

AI coding can help devs like you produce code faster, better, more useful than before…and code that's potentially much more dangerous, too. Technical debt. Wasted time, energy, and effort.

A tool like Claude Code can be your best friend, or the worst teammate you've ever had.

It's all in how you use it.

Most devs start AI coding the wrong way

In my experience, there are two main errors that most devs make with AI coding tools:

Both of these approaches are reactive, and you don't need decades of experience to know what happens when dev teams operate in reactive mode.

What you need is to design your own proactive, productive middle path. Neither YOLO nor OH NO.

A confident path, because you can trust your experience, your skills, your process, and your tools (in that order). In other words, the engineer's path.

When you take the engineer's path, you can get extraordinary results from Claude Code:

• faster development
• better automated tests
• higher quality software
• more fun, less grind

Best of all, you'll free up time to focus on your highest-value work: planning, designing, and actual engineering.

So, how can you thread that needle between anxiety and over-excitement and build that confidence?

First step: Analyze like an engineer

Production environments - especially existing codebases - require a professional approach.

Fancy tools can't eliminate engineering. In fact, the better the tools get, the wider the gulf gets between vibe coders and real engineers.

A real engineer knows the real work of software development has never been lines of code.

They know that engineering happens in the brain, not the IDE.

You, the engineer, have got to have a plan to prevent:

• Data loss or destruction
• Non-compliance with industry or corporate standards
• Breaking changes subtle enough to escape notice until it's too late
• Most existential of all: getting stuck in the mud as you lose your "sense" for a codebase

That last one should really be on ALL of our minds, because losing this sense isn't just an emotional reaction. It's a genuine cognitive loss that leads to a loss of understanding, team confusion, and the kind of overwhelm that does real (and possibly long term) damage to devs and their teams.

These are high stakes, and the costs can be staggering.

But these problems aren't new problems, even though they're accelerated by new tools.

Software developers like you have faced these same challenges with every innovation: from scripting languages to GUIs to advanced web frameworks, from outsourcing to distributed computing.

This is the legacy of engineering: our tools are always changing. And we're always adapting.

We can use these new tools to accelerate new solutions, too.

That's why real engineering skills matter, today, tomorrow, and forever.

Second step: Think like an engineer

A real engineer knows their tools intimately: what a tool is good for, how and where it excels, where it fails or falters, and how it breaks.

And, as an engineer, you should know that while AI is a new and unusual tool… it's still a tool. One you must fit intentionally into your professional software development process.

You can build the process that makes these tools trustworthy.

These are the core engineering skills you'll need to get the most out of Claude Code, or any AI-coding assistant:

• Communicating: clearly sharing your thoughts, ideas, observations, requirements
• Anticipating: quickly identifying potential problems before they happen
• Planning: reliably turning huge chunks of work into well-formatted PRDs
• Decomposing: thoughtfully breaking down big plans into well-scoped, iterative tasks
• Delegating: confidently knowing what & how to offload tasks
• Systematizing: consistently building feedback loops that produce quality outputs (types, tests, & reviews)
• Parallelizing: mindfully balancing quality assurance and coding, for maximum productivity

These aren't new ideas. This is engineering.

And I'll let you in on a little secret: these higher-order skills apply just as much to working effectively with people, not just coding agents.

Real engineering is the only way to generate safe, worthwhile, long-lasting code… by hand, or with AI tools.

If you don't use these skills - or you don't consistently use these skills - in your dev process today, today is the day to learn how.

And if you do use these skills today, writing code by hand, now's the time to learn how you can apply them with Claude Code.

The way you produce lines of code may change, but the core fundamentals of producing well-engineered software is eternal.

Your engineering skills make the difference between success and failure, enhanced productivity and technical debt at speed, safe code and code that risks everything.

Now: Learn to use AI coding like a real engineer

When you practice real engineering, you unlock the power of Claude Code.

For example, I've always dreamed about building the perfect course creator, designed specifically for my personal course development process.

From keeping my ideas organized, to the actual recording and editing process. I just knew that the right piece of software could let me focus more on making the learning experience amazing.

But this is my livelihood. It couldn't just be a vibe coded weekend project. I needed a reliable, professional grade tool, and knew that justifying that kind of time investment never made sense.

So for multiple years, that tool remained a pipe dream, with many aborted attempts.

But since I've started using Claude Code with Ralph loops, I've shipped over 1,000 commits, and created and closed over 500 of my own issues. Today, my professional-grade course creator includes things I never imagined possible:

• a full in-browser video editor
• AI writing assistant with 14 unique modes
• Dropbox/CMS/YouTube/X/LinkedIn sync

It’s all powered by TypeScript (obviously), and backed by the amazing Effect.ts library.

And here's the kicker: I built everything in my spare time, while running my business and enjoying life as the loving dad of my 2-year old.

I want to share that power with you.

So I went into brainstorm mode, and asked myself:

What do developers need to become engineers empowered by Claude Code?

And that's how AI Coding for Real Engineers came to be.

I designed a new 2-week course for you, where you will…

• learn & refresh timeless engineering skills
• update your engineering skills to include a new AI-assisted development workflow

You'll learn the fundamentals of how Claude Code works, what it can & can't do, how to prep your work, manage and steer it with token-efficient but reliable rules, keep your projects on track and on spec, test outputs, and ensure quality at every step.

You'll learn how you can fit Claude Code into a real-world production environment, with an existing codebase and procedures.

And you'll experience how correctly wielding a powerful tool like Claude Code can strengthen your abilities and deepen your understanding of your codebase.

In other words: You'll go from vibe coder to AI Hero.

(All these fundamentals apply to other AI coding tools, of course - but I've picked Claude Code because, for me, it's the current best-in-class.)

Join AI Coding for Real Engineers starting Monday June 1

My newest 2-week async cohort course offers you my best AI coding material to date, based on real-world production experience. And it's all designed for you to apply to your real-world projects.

Full Schedule: What you'll be learning, doing, and building each week

Each Monday, you'll get full access to that week's lessons, letting you pick your pace and schedule. Rip through lessons, or take your time; revisit to make sure you understand, and bring questions to Office Hours!

Here's what's you can look forward to, in detail:

Pre-course: Getting to Know Claude Code

To help you hit the ground running on Day 1, once you enroll you'll get immediate access to a collection of "Getting to Know Claude Code" lessons that I designed specifically for this course.

If you're brand new to Claude Code, this will help you get you comfortable with key terminology and features in the CLI coding assistant. And if you're already comfortable, this will be a useful and breezy refresher!

Week 1: Fundamentals, Steering & Planning

This week, we will build your engineering toolkit for working with an AI coding agent.

We'll start by making sure you understand the real constraints of LLMs, and how to manage them, all while using Claude Code to successfully build a new feature in a production-grade codebase.

You'll learn how context windows work, why managing them carefully matters, and how subagents can help you keep your context window clean. We'll practice exploring large codebases and using the Plan/Execute/Clear loop.

Combined, all of these techniques help your coding assistant stay in the "smart zone" where it writes great code, and avoid the "dumb zone" that sends it in circles.

Then you'll learn to steer the coding assistant. You'll write an AGENTS.md file that doesn't waste tokens on unnecessary instructions, and instead teach your agent how to conform to your team's preferences and expertise using custom skills and progressive disclosure (all while preventing wasted tokens).

By the end of the first week, you'll be planning and building features like a senior engineer:

• decomposing full-size features into context-window-sized chunks to build
• writing PRDs with a purpose-built skill
• using tracer bullets to validate architecture before committing to a full build
• and executing multi-phase plans that ship features far bigger than any single agent session can handle

We'll wrap up Week 1 with Office Hours. Bring your questions, get direct answers. Can't make it live? Submit your question in advance and get the recording and transcript.

Week 2: Feedback Loops, AFK Agents & Human-in-the-Loop Patterns

Week 2 stacks on top of everything you learn in week 1, introducing systems that make AI-assisted software engineering both trustworthy and reliable.

First, you'll find out why a green CI pipeline is non-negotiable when working with agents, and learning what kinds of tests do (and don't) belong in your agent's red-green-refactor workflow. You'll build a custom "skill" that encodes your team's quality bar into every task, and pre-commit hooks that add a deterministic safety net to your code quality!

Then, the key to automating your engineering workflow: the "Ralph Wiggum" loop, a deceptively simple and incredibly powerful pattern for letting your agent autonomously build entire features into your app.

You'll learn and practice running Ralph in two modes: first, with "human-in-the-loop" mode (hint: you're that human). HITL gives you both visibility and control while you're learning.

Then we'll take off the training wheels and set up a safe sandbox for Claude Code to operate entirely on its own (while you are AFK), pulling from a GitHub Issues backlog, prioritizing fixes and features, and committing changes to branches. With just a bit of practice, you'll have it churning through work while you focus on other things that matter most.

By the end of week two, you'll have a clear sense of when to stay in the loop vs. go AFK autonomous, including:

• how and when to run research phases that save future sessions hours of work
• the power of flexible kanban boards instead of upfront plans
• and maybe most importantly, how to make sure that your codebase is prepared for AI agents to actually navigate it effectively.

Week 2 Office Hours with Matt. Same format, deeper questions. By now you'll have real experience to draw from, so we can help you connect the dots between your practice and real work in your codebase!

Your future with AI coding is inside

This course focuses on the Claude Code CLI, which is where I've honed these skills and gotten the best results myself. So it's the best way for me to teach you!

But once you've practiced the skills here, you'll be able to use the same techniques and knowledge in just about any CLI coding assistant, with virtually any frontier AI model.

And the more you practice these skills, the better judgement you'll bring to using coding assistants in production-grade codebases.

Learn on your schedule, from anywhere

AI Coding for Real Engineers is the most flexible course I've ever offered:

Instead of being yoked to a semi-daily schedule, you'll receive the entire weeks' worth of lessons each Monday, so you can work ahead or take it slow.

You'll gain access to 6 live Office Hours across 3 days - staggered for different time zones - to give you the most opportunity to get personal help directly from me. Ask ahead if you can't make it; recordings and transcripts are always included.

And, of course, you'll get invited to my private AI Hero Discord, with an exclusive channel for you to meet & learn along with your fellow attendees.

This way, you get the best of sync and async courses in one great experience.

What's included

For just $995 $696 - save 30% during launch week! you'll learn everything you need to use AI coding like a real engineer.

Here's what you get when you enroll:

• All lessons and coding exercises. Tight, snappy videos and hands-on practice in a safe sandbox environment. Yours to keep forever.
• The course code repo. Clone it, reference it, build on it. It's yours.
• Pre-course access. Getting to Know Claude Code lessons available before the cohort so you are ready on Day 1.
• Live Office Hours with Matt. Live Q&A sessions on Monday June 1, Monday June 8, and Friday June 12, with two time-zone friendly options each office hours day. Submit questions in advance if you can't join live.
• All recordings and transcripts. Every office hours session, captured and available for replay.
• Exclusive Discord access. Learn alongside your fellow students, ask questions, share wins, and soak up the energy of a group of smart folks figuring things out together.

Everything listed here is yours to keep so you can revisit them any time as a reference, as a refresher, or to go deeper with exercises you didn't get to during the cohort.

Enroll today to gain the skills, workflows, and confidence to use AI coding tools the way they were meant to be used: as a force multiplier for real engineering.

Frequently Asked Questions

Which Claude plan do I need?

Claude Pro (\0/month) works - it includes Claude Code and all models including Opus. For the best experience, especially for longer autonomous sessions, Max 5x (\00/month) gives you 5x the usage limits and priority access.

Can I use a different AI coding tool?

You can use ANY tool you prefer - the engineering principles transfer to Copilot, Cursor, Windsurf, and others. Matt teaches with Claude Code. We recommend using Claude for the course, then applying what you learn to your preferred tool.

Do I keep access forever?

Yes - all lessons, exercises, the code repo, office hours recordings, and transcripts are yours to keep.

What if I can't attend live or have timezone conflicts?

Office hours are scheduled at morning and evening times (Matt's timezone) for global coverage. All sessions are recorded with transcripts. You can submit questions ahead of time. All lessons drop Monday each week - work at your own pace.

What's your refund policy?

30-day refund policy. Contact team@aihero.dev.

More questions?

For pricing, PPP, team seats, invoices, gifting, and platform questions, see aihero.dev/faq. Or email team@aihero.dev.