Rather than expressing my bias towards them first , check out these threads by other people:
https://community.meraki.com/t5/Switching/MS-390-when-is-it-ready-for-action/td-p/80612
And you mostly see mosts from unhappy people who bought them:
https://community.meraki.com/t5/Switching/MS390-DO-NOT-BUY/m-p/95581
https://community.meraki.com/t5/Switching/MS390-LLDP-topology-issues/m-p/92618
Back to my bias again. I'm also a Cisco Partner. I would not put an MS390 into a client at this point in time. Maybe in a couple of years, I'll re-look at it. I think it will take this long to get on top of the bugs.
I don't want the damage to our reputation by installing core infrastructure like switching that does not operate 100% reliably.
Based on some comments you have made (I have made a guess about the size of your network); I typically use MS425-16's for network cores or server access blocks. They are an all 10Gbe SFP+ switch.
I'm also a big fan of the MS325-24X series for mid-size network cores. Each one has 24 copper ports, 8 of which are MGig, and 4 x SFP+ ports. You can stack them together and get a really good density of Gigibat, MGig/10Gbe copper and SFP+ ports.
For the access layer blocks, you could also look at the MS355 series (lots of MGig if that is important to you).
I mostly use MS225's (which have built-in 10Gbe SFP+ uplinks) which I 10Gbe connect back to the MS425's.