A good schedule is one you can hold for three months. Everything else is secondary: a perfect plan abandoned after two weeks is worse than a modest one that survives.
Start from the volume you can carry
Count hours, not posts. One teardown is an hour and a half with edits. Five a week is a working day. If you do not have a working day for the channel, there will not be five teardowns a week.
An honest start is two or three posts a week. Scaling up is easier than recovering from a collapse.
Days
Even intervals read better than three posts together and silence. Monday, Wednesday, Friday works for almost any topic. Weekends are usually weaker, though it depends: for entertainment channels the opposite.
Times
Rough guides are the morning commute, lunch and after eight in the evening. But a guide is not a test — your audience may differ. On finding yours, the best time to post.
A buffer
Keep a week queued. Not for resilience but for calm: with an empty queue, any disruption to your plans becomes a gap.
Testing the schedule
After a month, look at reach by day and time. If the spread is real, move one parameter at a time. And do not change the schedule more than monthly: an audience takes longer than that to learn a rhythm.
How Neurogram does this
Everything above is manual work that runs into the number of hours in a day. Neurogram does it on your own accounts and without you: it collects channels and audiences, writes comments and replies with a language model, and keeps limits and quiet hours so the accounts survive. See pricing or look at your case.