Claude Code worked for 20 or 30 minutes in total, and produced a Z80 emulator that was able to pass ZEXDOC and ZEXALL, in 1200 lines of very readable and well commented C code (1800 lines with comments and blank spaces). The agent was prompted zero times during the implementation, it acted absolutely alone. It never accessed the internet, and the process it used to implement the emulator was of continuous testing, interacting with the CP/M binaries implementing the ZEXDOC and ZEXALL, writing just the CP/M syscalls needed to produce the output on the screen. Multiple times it also used the Spectrum ROM and other binaries that were available, or binaries it created from scratch to see if the emulator was working correctly. In short: the implementation was performed in a very similar way to how a human programmer would do it, and not outputting a complete implementation from scratch “uncompressing” it from the weights. Instead, different classes of instructions were implemented incrementally, and there were bugs that were fixed via integration tests, debugging sessions, dumps, printf calls, and so forth.
New methods boost reasoning in small and large language models，更多细节参见新收录的资料
，这一点在新收录的资料中也有详细论述
Гуменник рассказал о переживаниях перед финалом Гран-при России17:42
During development I encountered a caveat: Opus 4.5 can’t test or view a terminal output, especially one with unusual functional requirements. But despite being blind, it knew enough about the ratatui terminal framework to implement whatever UI changes I asked. There were a large number of UI bugs that likely were caused by Opus’s inability to create test cases, namely failures to account for scroll offsets resulting in incorrect click locations. As someone who spent 5 years as a black box Software QA Engineer who was unable to review the underlying code, this situation was my specialty. I put my QA skills to work by messing around with miditui, told Opus any errors with occasionally a screenshot, and it was able to fix them easily. I do not believe that these bugs are inherently due to LLM agents being better or worse than humans as humans are most definitely capable of making the same mistakes. Even though I myself am adept at finding the bugs and offering solutions, I don’t believe that I would inherently avoid causing similar bugs were I to code such an interactive app without AI assistance: QA brain is different from software engineering brain.，更多细节参见新收录的资料