The values Cursor actually screens

Once your cost model and your GPU-economics answers are settled, what's left is temperament. Would you walk into a room with the ML lead and the CFO, say the inference margin is worse than anyone is admitting and hold that line on the data when someone senior pushes back?

Cursor publishes no interview stages. The posting describes a very flat, small, talent-dense team, so prepare as if a senior leader or founder may be in the room. That is a general-industry inference, not a published fact. Nobody is checking whether you can run a standup. They are checking whether you'd be a trusted thought partner on inference cost and capacity decisions without a management layer to hide behind.