1 min lesson
Pressure-test every idea against four constraints
Match each case in "Pressure-test every idea against four constraints" to the signal and response that fit it.
Step 1 of 3
Pressure-test every idea against four constraints
Does it stay inside the surface's latency budget?
Could it ever be confidently wrong in a way that costs more than it helps?
What does it cost per request at scale and does it degrade gracefully?
Does it serve individuals, enterprise or both, and is that the right bet for now?
Learn more
Full explanation
Bucket three: what you would not build
Bucket three: what you would NOT build
The strongest signal of taste is a confident cut. Naming something you'd explicitly decline, and why, proves you can set direction, which is the actual job.
Have one rehearsed "I would not build this" with a real reason: a flashy feature that adds latency to Tab, a consumer-style growth gimmick that doesn't fit a developer audience or an enterprise control that would slow the individual experience. Pair the cut with the tradeoff it protects. "I'd cut X to defend Tab's latency budget" shows judgment; "we should build everything" shows none.
Before you present any idea, run it through the role's one question: does this make Cursor something a developer reaches for every single day? A paper-cut fix passes by removing daily friction. A bigger bet passes by deepening a daily habit. If an idea only matters in a quarterly review, it's probably not the highest-impact thing to ship now.
Learn more
Optional practice
Test yourself on Pressure-test every idea against four constraints
QIn the "discuss ideas" round you're asked for your top bet for Cursor. What structure makes the answer land?
Learn more
Optional practice