The usual for me. I got rapidly 2 off. From there, I tried variations until one cut 2 steps. But I wasn't able to "see" the one cutting the 2 steps. It just works eventually.
Side note: for me, exit was straight clear, thanks to obvious "impossibility" for 3 out of 4. :D
The amount of steps consumed just to leave the start zone gave me the key to the solution immediately. If there would be an ambiguity about the side to work on, it would make it a bit more challenging maybe?
The configuration felt therefore unbalanced to me.
For me, pipelining remains a matter of luck. I know all the tricks, though, where to apply them, at what location and in what order isn't something I'm able to plan. I just try one variation after another and eventually one wins. It's very rare that my intuitive arrangement is straight the winning one.