2023 HowIWonSingaporesGPT4PromptEngi

Subject Headings: Prompt Engineering Competition.

Notes

It describes Sheila Teo's experience winning Singapore’s first GPT-4 Prompt Engineering competition, over 400 participants.
It emphasizes the role of prompt engineering as a blend of art and science, combining technical knowledge, creativity, and strategic thinking.
It introduces the CO-STAR framework, developed by GovTech Singapore's team, for structuring prompts to optimize LLM responses.
It discusses using delimiters in prompts, such as particular characters or XML tags, to help LLMs understand and segment different parts of the prompt.
It covers the concept of System Prompts in LLMs like ChatGPT, explaining how they act as a filter for consistent application of user instructions in a chat.
It explores the potential of using LLMs for dataset analysis, emphasizing their strengths in pattern recognition and trend analysis.
It acknowledges LLMs' limitations in tasks requiring precise mathematical calculations, suggesting using plugins or code for such analyses.
It provides a real-world example of analyzing a Kaggle dataset using GPT-4, illustrating the practical application of LLMs in data analytics.
It concludes with Sheila Teo's reflections on the competition and her offer to assist others in prompt engineering, indicating a community-focused knowledge-sharing approach.

A deep dive into the strategies I learned for harnessing the power of Large Language Models (LLMs)

;

	Author	volume	Date Value	title	type	journal	titleUrl	doi	note	year
2023 HowIWonSingaporesGPT4PromptEngi	Sheila Teo			How I Won Singaporeâs GPT-4 Prompt Engineering Competition						2023