Skip to content

DCS-training/Prompt-Tips-GenAI

Repository files navigation

Prompt-Tips-GenAI

This is the Efficient Prompting for Generative AI Tools training course material repository

Check out training course slides: Efficient Prompting Training

General Structure

Course is designed for giving a general understanding of the efficient use of Gen-AI tools focusing on the importance of Prompting (Prompt Engineering). Starting with a brief introduction of Gen-AI tools for practical reasons, different prompt strategies will be discussed by doing some interactive examples. The content of the workshop will cover the concepts of ‘Introduction to Gen-AI tools’, ‘Prompt Engineering’, ‘Prompt formulation’ and ‘Effective Prompting’. Additionally, certain plug-ins in ChatGPT with Data Analytics feature will be explored with good prompt practices. Course is aimed to be delivered in a 2-hour session with certain hands-on components and interactions with audience.

Main Course sources

Course slides are prepared as a mixture of different related sources but mainly focusing on the followings;

Prompt engineering by OpenAI

Prompt Engineering Guide

Acknowledgements

Further thanks for inspiring people including but not limited to Andrew NG, Sebastian Raschka.

Related keywords

Generative Artificial Intelligence (Gen-AI), Large Language Models (LLMs), ChatGPT


Part 1: Introduction to Gen-AI and Prompt Engineering

  • Welcome and Overview
  • What are Gen-AI Tools or LLMs?
  • Importance of LLM Settings
  • Basics and elements of Prompting.
  • Key Components of Prompt Engineering

Short Break

Part 2: Techniques in Prompt Engineering

  • General Tips for Designing Prompts
  • Strategies for Effective Prompting
  • Group hands-on: Prompt Formulation Practice
  • Techniques in Advanced Prompt Formulation

Part 3: Exploring ChatGPT with Data Analytics

  • Brief Introduction to ChatGPT Plugins for Data Analytics

Closing

  • Feedback and Closing Remarks

Additional Notes

  • Hands-on components are included in various parts of the session.
  • The agenda is subject to slight modifications to better suit the pace and interaction level of participants.
  • User Preferences on Related Tools:
    • The course will include demonstrations and use cases involving popular LLMs such as ChatGPT or its competitors. For the participants who are not able to use recent version of ChatGPT, there will be open-source platform suggestions to use
    • Disclaimer: Different tools can require different prompt strategies in general!