So you are trying to teach a large online course. Everyone knows this will be boring and monotonous, right? What if we told you that our large, online, livestreamed courses feel more interactive than our old small in-person courses? Part of that is due to the parallel chat, but a significant part also comes from co-teaching, the topic of this article.
The basic idea is that we accept that the audience is probably going to be quiet, and plant someone to interact with the instructor. This person doesn't pretend to be a student, but directly acts as a co-instructor and the course becomes a discussion between them. This is different than having two instructors who alternate teaching, or an instructor or helper. It's more like a pilot-co-pilot situation, where both pilots have a certain focus, but the flying is a continual team process with each having a focus (flying and monitoring) but both constantly providing information to each other.
Doing co-teaching in practice requires some care, but isn't that hard. There are different models, usually we try to designate a primary who is going over the course material, and the other co-instructor acts as a learner, asking questions and engaging by being the "voice of the audience". The co-instructor who isn't most actively talking spends some time watching the parallel chat and raising these questions. During demo times, one common strategy is that one person is guiding and explaining the big picture and the other person is typing and explaining the small picture. Whatever happens, it works. And usually very well.
Besides the much greater interaction, there are other benefits. It's much easier to onboard a new instructor - one can almost go straight from advanced learner to co-instructor, since "asking questions a learner might have" is enough to start. Co-teaching also reduces the stress of preparation: the instructors still have to prepare, but with two people, by simply pausing or asking a question to the other person, any gaps can be filled in.
Perhaps the biggest disadvantages are the need for more people and to coordinate. But, there is probably less preparation total needed. It requires more staff, but we need a continual flow of instructors coming in anyway - so this helps us long-term.
What's the minimum size course where this makes sense? In theory, it could work for very small courses, but at that size you would probably hope that the audience interacts directly. At a few tens of students, it might work, if you can keep the co-instructor discipline working well - but that might easily be forgotten if there are many questions from the audience. But I personally think that even small courses benefit from two brains, and it worth trying co-teaching at any size.
Overall, co-teaching has revolutionized our teaching: we can feel more interactive in large courses, we can bring in new instructors more quickly, and we can teach better. It really seems to solve the boring "500-person lecture" problems that I had when I was in university. We now use co-teaching for anything that needs to seem "professional", from 20-person instructor training via Zoom to our 500-person livestreamed Python for Scientific Computing courses.