What is the general design of these new math solving systems? [D]
The user inquired about the general design of a new mathematical solving system. The existing system allows models (often Aster) to generate LEAN statements and submit them for compiler inspection. After successful compilation, the statements are added as facts, and the process ends once the complete proof of successful compilation is achieved. Some papers are hundreds of pages long, suggesting that the proof might be constructed in parts, assembled, and then submitted. The user plans to implement a self-developed version to verify doubts regarding high-dimensional geometry and seeks advice on methods for local combinatorial macro concepts, unpublished ideas, and hardware requirements.