Make rendering (delivery) jobs smarter

I think it is common sense that job processing is intended for long-running jobs that work unattended.
Unfortunately Jobs in Resolve are not good for typical batch processing, because of these issues:

• If there is a problem with a single clip, the whole job aborts. So you have to attend the processing periodically to verify that it's still running. That is bad. Probably it would be better if the job continued and created a summary of problems at the end.
  
• Even when rendering clips as individual files, it is not possible to restart rendering where the error was; instead the whole jobs starts at the beginning.

At work I have backup software that can backup multiple servers with multiple filesystem objects in one job. If occasionally some server has a problem (the jobs run unattended at night), I have an option "restart failed objects" the next day, and then another job is created automatically that backs up just those objects that should have been backed up, but were not (due to errors).
That's exactly what I'd like to see in Resolve. And of course the backup job continues when it sees some error...