# Expert Role Prompting: How to Get Better Answers From ChatGPT and Claude

> There is a pretty simple way to get better answers out of Claude / ChatGPT. I first got this idea from a Karpathy post, since then I use this approach daily.

If you ask a language model a question, you'll get a prediction of the average of the internet's answer - whatever is said most often in reply to a question like yours.

You improve the output with one additional step: you force the model to first reason about which **specific person** would be best to answer this question, and then answer as them.

This works so well because experts in each field usually have a large public body of writing: books, lecture transcripts, blogs, etc. that went into the model's training data, so the model has a great understanding of how an expert thinks - and simulating "what does Paul Graham think about my startup idea" pulls far better answers than the crowd average.

I initially got the idea for this approach from [this](https://x.com/karpathy/status/1997731268969304070?lang=en) Karpathy tweet:

![Andrej Karpathy's post on framing prompts around which expert would answer](https://lukasvonkunhardt.com/_assets/karpathy_expert_framing.png)

Since then I have experimented with it and found that the best way to use this framing is to have the model really pick one **specific** person, or a panel of specific people to discuss this, and first explain **why** the chosen people are particularly well suited to answer this question. This seems to reinforce the character simulation.

I turned this into a reusable skill. Install it with one command, `npx skills add lukaskunhardt/skills`, or grab it on [GitHub](https://github.com/lukaskunhardt/skills). The same repo has the [LLM wiki skills](https://lukasvonkunhardt.com/my-favorite-memory-system-for-llms/) I use for project memory.

---

Source: https://lukasvonkunhardt.com/the-simplest-way-to-improve-llm-answers/ · Author: Lukas von Kunhardt · Published: 2026-05-30 · Updated: 2026-08-20