Interview knowledge
Article

LLMs reward expertise

Read the original →

Highlights

Passages saved from the original.

  • In the 2010s, if you had technical gaps (say, you couldn’t write CSS), you had to either rely on a skilled colleague or just hope that the answer to your exact problem was out there on the internet. Today, everyone can write sort-of-okay CSS by delegating the task to an LLM. LLMs make everybody into a generalist
  • The most important skill in prompting is expertise in the domain you’re prompting for.
    • By signalling expertise, Tao shunts the model into “talking-to-mathematicians” mode, not “explaining-to-amateurs” mode
  • if you have domain knowledge, you can wring far more value out of the same LLM by steering it hard in the direction you want. Most of us will have to do a mix of both these approaches, since we have domain knowledge in some areas but not others.
  • the human is the bottleneck, not the model, because the difficult part is in communicating to the model exactly what kind of solution the human wants

Summary AI-generated

Summarised by an AI assistant when this source was saved. Double-check against the original.

In summary, the most interesting aspect is that domain expertise, rather than prompt engineering tricks, is the true differentiator for effectively leveraging LLMs, as experts can accurately evaluate outputs, steer complex reasoning, and extract high-value insights that novices miss.

  • Domain knowledge outweighs generic prompt-crafting techniques in determining LLM effectiveness.
  • Experts can quickly identify hallucinations, steer conversations precisely, and push boundaries that novices cannot.
  • Concrete familiarity with a specific domain yields more value than abstract engineering principles.
  • Human expertise remains the bottleneck for high-quality AI output, as models already contain the information but require knowledgeable users to extract it.

Also cited by