Answering questions in your own words is good for learning. Answering them by clicking a button is better than not answering them at all.

That sounds like a low bar, but it’s actually the insight that shaped the quiz system. At 20:00 after a full day I’m not going to write paragraphs about container networking from memory. But I will click through ten multiple choice questions if the friction is low enough and the questions are good.

The Problem with Manual Questions

The DevOps Learning Command Center already had three questions per lesson, generated in advance and stored in the pre-built lesson files and Google Calendar events. They were good questions — I wrote them carefully based on the lesson content.

But they were static. The same three questions every time, typed into a calendar event, answered (or not) by looking at the event description on your phone. No feedback. No score. No sense of which concepts had stuck and which hadn’t.

What I wanted was something closer to an actual test.

Quiz Generator Plugin

Obsidian has a plugin called Quiz Generator that uses AI models to generate interactive exam-style questions from your notes. Multiple choice, true/false, fill in the blank — rendered in a proper quiz UI with real-time feedback. It supports Anthropic as the AI backend, which matters because Claude generates contextually accurate questions for technical content.

The workflow is: point it at a note, generate questions, answer them by clicking, see your results. Done.

The /make-quiz Skill

The interesting part is what note you point it at. Today’s lesson alone gives you three or four decent questions. But the spaced repetition system schedules past lesson reviews on specific days — why not include those lessons too?

/make-quiz <N> assembles a single quiz material file before the 20:00 session:

  1. Today’s lesson — sub-lessons and questions from _Learning Plan/
  2. Scheduled reviews — past lessons due for review today, pulled from the calendar note and their session notes in _Session Notes/
  3. Two random past lessons — randomly picked from all completed session notes

The random picks are the part I’m most interested in. Spaced repetition handles the structured review. The random picks handle the long tail — older lessons that aren’t on the formal schedule anymore but should still be in memory. For someone learning at 56, the research on this is clear: varied retrieval practice beats blocked practice every time.

The assembled note goes to _Session Notes/YYYY-MM-DD — Quiz.md. Open it in Obsidian, run Quiz Generator, answer 8–12 questions covering up to five lessons. Takes about fifteen minutes.

What Gets Flagged

If a session note has #revisit tags — concepts I flagged during the lesson as confusing — those surface in the quiz material with a warning callout. Quiz Generator sees them, weights the questions accordingly, and I end up answering more questions about the things I found hard rather than the things I already understood.

The whole loop closes: lesson → session note → difficulty rating → spaced repetition schedule → quiz assembly → multiple choice questions → #revisit flagging → next quiz. Every piece feeds the next.

Whether it actually makes me a better DevOps engineer is a question I’ll be able to answer in about five weeks.