It is such a relief to have him all completed and know that he works like I want him to. Everything he does is programmable. That is, I can adjust how long the lights are on and off, how fast and how far he moves forward and backward. But it's all programmed BEFORE the show; sort of like how I memorize lines, RD-3D will have "memorized" things he will do.
For example, I'll ask him questions and the green light will represent "yes" and the red light will be "no." But he won't actually hear the questions, just wait a set amount of time and then turn on the green light for a second. Also, the shaking back and forth he does near the end of the video I think I'm going to use as a joke in the show. I'll say something like "Okay, we have an agreement. Let's shake on it," and he will begin to shake.