If you are tuning this with your tuning company, I'm assuming they datalogged the before and after data values.

What psi were you achieving before vs now > What lambda was achieved before vs now? > What ignition timing was achieved before vs now?

If you can answer those questions, it would be much easier to diagnose what has been missed in the file edit OR what hardware needs to be repaired/resolved/upgraded..etc.

 

Of course, car was logged. However it will not be that easy. I changed lower pulley last year and tune was done this year. So no data log on previous "healthy" setup.

With lower pulley and without tune adjustments car had very rich AFR (more like 12 vs 14.7). Psi was already changed as per pulley - if I recall properly 0.8 to 0.35 (in the peak). Which makes me feel that boost is "lost" somewhere (as per MoscowLeaper suggestion throttle plate in S/C). Now AFR is healthy. Don't recall values for ignition timing.

PS. I have water/meth injection (before throttle body). But did not have any issues with MAF sensor etc. So should be okay.

 

It may be a PCM closing a throttle, since the target torque/boost are acieved and/or too high boost is detected. Too many variables to guess, but the value of total airflow may give a hint.
Also, there's a hardcoded limit of 1024Nm total (output torque+torque losses), which is a bit tricky to overcome