# Claude and ChatGPT for Ad-Hoc Tasks, a Case Study - Simon Willison Synced: [[2024_03_27]] 11:06 AM Last Highlighted: [[2024_03_27]]  Summary: The document is a case study showcasing the use of Claude and ChatGPT for an ad-hoc task. It demonstrates how the author utilized these tools to convert a shapefile of the Adirondack Park boundary into GeoJSON format efficiently. By leveraging the capabilities of these AI models and external tools like geojson.io, the author successfully refined the polygon representation of the park, highlighting the importance of domain knowledge and experience in driving LLMs effectively. The study emphasizes the practical utility and impact of LLMs in enhancing productivity and problem-solving, despite the complexities involved in optimizing their usage. ## Highlights [[2024_03_27]] [View Highlight](https://read.readwise.io/read/01hszhqf4hkwm1kz1ceyjebknk) > If you have the right combination of domain knowledge and hard-won experience driving LLMs, you can *fly* with these things [[2024_03_27]] [View Highlight](https://read.readwise.io/read/01hszhrmw0e576c2ex4652nvhz) > One of the greatest misconceptions concerning LLMs is the idea that they are easy to use. They really aren’t: getting great results out of them requires a great deal of experience and hard-fought intuition, combined with deep domain knowledge of the problem you are applying them to.