This looks pretty neat! I had a few questions:

- Is there any way to add `perf` profiling?

- Could allocation graphs be added, similar to what pprof offers (e.g. the one midway through this article https://blog.detectify.com/2019/09/05/how-we-tracked-down-a-...)? I've found these to be very helpful in practice since flame graphs make it harder to see what lower-level functions are being called a lot.