true - but the way LLMs are trained, google's RLVR is different from anthropic's is different from openai's. you'll get very good results sending the same 'revi

true - but the way LLMs are trained, google's RLVR is different from anthropic's is different from openai's. you'll get very good results sending the same 'review this change' prompt (literally) to different models.