Good post, but I think it’s worth setting realistic expectations on feature release.
My suggestion is to plan for the release of Real time API by planning and having your workflow down beforehand, this way you’re in prime position to take advantage of its benefits.
I’d rather wait and have a robust tool instead of getting access and fumbling with bugs and issues while they’re still being ironed out.
Look up @ConicuCallum on Twitter to see how I’m planning to utilise real time API.
If you’ve got a team who knows what they’re doing, you’ll be in prime position to take advantage of the features when OpenAI have decided it’s time to release.
Hope this helps.