Over 40,000 engineers learn with The AI Merge. NVIDIA creator partner since 2025.
START HERE
Where AI work usually gets stuck
Usually in one of three places. Start with the one closest to the problem on your desk this week.
Who The AI Merge is for
No research background needed. If you build and run software, you already have most of what it takes. What you won’t find here: prompt tricks, tool roundups or hype.
BACKEND & FULL-STACK
Someone just asked you to “add AI” to the product.
Whether you’ve used AI before or not, it doesn’t have to be a black box. Once you see how a model really works, building with it stops feeling like guesswork.
SENIOR & STAFF ENGINEERS
You’re the one who has to make the call.
First question: do you need AI at all? If you do, which kind? Hosted API or your own model, cloud or on-device, managed or self-run. I show how to weigh each option before you commit.
MOVING INTO AI ENGINEERING
You want AI to be your actual job.
This is the order I’d learn it in if I were starting again: fundamentals first, then engineering with AI, then whole systems, with something to build at each step.
TECH LEADS
Your team is shipping code written with agents.
Every line an agent adds is a line your team has to keep running. Coding with AI only pays off when it builds on the practices you already have: standards, reviews, tests, and what your team already knows.
WHY THE AI MERGE
Where software engineering
and AI merge.
HOW I TEACH
See the whole system first
Every topic starts with a diagram of the full system, so you know where a piece fits before going deep on it. It’s how I learn too: if I can’t draw it, I don’t understand it yet.
Weigh the trade-offs
For each part you see the real options, what each one costs and how it breaks. You get the reasoning behind a choice, so you can make a different one for your own setup.
Then build it, with AI
Only then do we build, with coding agents used the way I use them at work: one piece at a time, the design as context, every change reviewed.
Courses
The first is The AI Atlas. A hands-on course on vision and agent systems comes next.
A few pages from inside the Atlas.
Sample pages from the Atlas. Use the arrows or swipe.
Two ways to hear from me.
FIELD NOTES
What I’ve shared lately
Everything else is on Substack
Things people ask me
Got a question that isn’t here? Reply to any Field Notes email. I read all of them.
What is The AI Merge?
An education brand for software engineers working with AI. Articles and courses on how AI works, engineering with it, and running AI systems in production.
Who is it for?
Engineers who already ship software and now work with AI: backend and full-stack engineers, senior engineers and tech leads.
Do I need a machine learning background?
No. If you’re comfortable with code and systems, you have what you need. The fundamentals are explained with an engineer’s questions in mind.
How do you approach a topic?
I map the whole system and its trade-offs first, then build. AI comes in once the design is clear. The guides follow the same order.
Where should I start?
Fundamentals if models feel like a black box, Engineering if you code with agents, Systems if you’re shipping AI features. The free articles are the easiest way in.
What’s free and what’s paid?
Articles and the newsletter are free. Courses are paid, one-time purchases with no subscription.
What is The AI Atlas?
The first AI Merge course: 240+ slides on how modern AI systems work, from hardware up. Join the waitlist and you’ll hear when it’s out.
Can we work together?
Yes. I’m an NVIDIA creator partner and take sponsorships from AI companies whose tools fit what I teach, as well as collaborations and talks. See the Work with me page.
