I've just added some very dodgy ALSA sound support to my Linux/xcb platform layer:
https://github.com/nxsy/xcb_handmade
This is the first time I've tried anything to do with sound at this level, so I'm just surprised I got something that mostly works most of the time. It has buffer underruns and more latency than I would like.
It would be awesome if someone with ALSA sound experience could help out - or even just someone interested in learning about it in the deep end.
Thanks!