I recently moved my team in the Cursor world and the first thing we learnt is "build as many custom rules as possible, including a rule to stop and make new rules". Cursor is an absolutely brilliant tool, productivity has gone through the roof, but only after we tamed it. You really have to give it small tasks for it to be effective and limit the scope of what it can affect so it doesn't run wild and try to refactor your entire codebase.
This hits hard — efficiency over headcount is the new rulebook. AI’s power is real but without deep context, it’s just a faster autopilot for distractions. Smart teams build with purpose, not just velocity.
I recently moved my team in the Cursor world and the first thing we learnt is "build as many custom rules as possible, including a rule to stop and make new rules". Cursor is an absolutely brilliant tool, productivity has gone through the roof, but only after we tamed it. You really have to give it small tasks for it to be effective and limit the scope of what it can affect so it doesn't run wild and try to refactor your entire codebase.
It would be nice if the world worked that way, but sadly speculative value is more important than real value
but there has to be a way to measure value in some way, maybe not the full spectrum of it, I hope my question is coming from the right frame.
This hits hard — efficiency over headcount is the new rulebook. AI’s power is real but without deep context, it’s just a faster autopilot for distractions. Smart teams build with purpose, not just velocity.
I analyse market using cycles – strategymaster.substack.com