Dr. Eliza is an AI engine designed to solve difficult problems, stated in natural language, in many different problem domains. While this may at first sound like wildly futuristic SciFi, its internal workings are surprisingly straightforward. Note that Dr. Eliza does NOT work particularly well for simple problems, and doesn't work at all for questions. In short, it makes crucial assumptions based on a presumption of difficulty.
Dr. Eliza relies on some basic principles of difficult problems, the most important being that only a tiny fraction of all problems ever rise to the point where anyone is going to document them for someone else to assist. In almost any given problem domain, there are only dozens of common problematical components of difficult problems, and not the thousands that you might first think. Of course, these components can be combined in thousands of potential ways, but Dr. Eliza carefully works with problem components rather than the much larger number of potential combinations of components.
Interacting with Dr. Eliza is a process rather than an act. In response to your problem, Dr. Eliza will probably ask you some questions. For the best results, instead of answering them directly, edit your problem statement to incorporate the answers to the questions.
Of course, if Dr. Eliza fails to solve your problem, your next stop is probably a professional person to help you. To minimize the cost and maximize the efficiency of that professional, you should start with a problem statement with all of the usual questions already answered. Hence, even when Dr. Eliza fails to solve your problem, it does help you wring out your problem statement for your human professional.
To use Dr. Eliza, describe your problem to Dr. Eliza. The first line should be the greeting, e.g. "Dr. Eliza," and the last line should be your name or pseudonym, and in between should be a detailed description of your problem. See the How to use Dr. Eliza section for more details.