LLMs reward expertise
Summary
The piece argues that domain expertise is the key to effective prompting of LLMs, not generic prompting tricks. It uses Terence Tao's math conversation as a case study to show how domain knowledge and critical questioning lead to higher-quality AI outputs.