The weekly letter from Systems Thinking Lab: systems thinking insights for junior engineers, framed through the seven building blocks.
You ask the AI to build the feature. It writes something that looks right, the checks you thought to run pass, and you ship it before lunch. For a while, that feels like winning.
Then production breaks at 2 AM, or code review asks why you built it this way, or an interviewer asks you to walk through a project with your name on it, and you go quiet. The code is yours. The understanding never was, because you were never really in the loop while it got built.
That gap is not a discipline problem. It is the shape of the job right now, whatever your title says.
If you are building software with AI, you are doing a senior engineering job. Not eventually. Now. An AI that builds in seconds still leaves the deciding to you: breaking the work down, questioning the plan before it builds, judging whether what comes back is good enough to ship. Those are the calls a senior engineer makes every day, and you are making them whether anyone told you or not.
Picture a senior at the whiteboard before a single line gets written. She does not ask, can this be built? She asks, what is actually being decided here? And how would we know this plan is wrong before it ships? AI changed what gets typed. It did not change who has to ask that question.
One move catches most of what goes wrong: before you let an agent build anything, ask it what it thinks. Then make it prove it. A weak answer just restates the plan. A real one points at the exact thing that would pass or fail. Ask what happens when two people hit this in the same second. A weak plan says it handles that. A real one names the exact row both writes land on. It costs you almost nothing. Skip it enough times, and you get the 2 AM version: fast code, no debugging trail, no answer when someone asks why.
I built a course around that one move. I think it is the actual skill gap right now, not typing speed. Nobody had put it in one place. It teaches you to command an AI agent the way a senior engineer would. That is how you code. The four courses after it, I through IV, build the judgment to know whether what it built is any good. That is how you grow. It is called Course 0, and it opens today. That is why this letter is arriving on a Sunday instead of the usual Saturday.
Self-paced: eight lessons and two hands-on labs where you run the method yourself instead of watching someone else run it. Ten short written questions close it out: feedback on your answers, not just a score, and unlimited retakes. Then three challenges on a small chat app you did not write, real code, real seams. A local chatbot you take over and make your own, real enough to break and rebuild: you complete two of the seven building blocks and work inside two more that are already running.
It is built for wherever you actually are. The one who ships fast with AI and cannot explain how it works when asked. The one who checks every line by hand, and it is starting to feel like a fight instead of craft. The one who plans well with the AI and loses everything the moment the context clears. And the vibe coder, still building one specific rep: sitting in the loop and pushing back on a plan before it builds.
Course 0 is $99, one time. Courses I through IV are $299, one bundle, open to anyone.
The seven building blocks stay free, same as always. Course 0 is where they stop being something you read and start being something your hands do.
P.S. Course 0 is open. If you want the details, they are here: https://systemthinkinglab.ai/course-0