Welcome to my latest video, where I share my experience working with the OpenAI API, based on a short course by Deeplearning.AI that I recently completed. The course focused on prompt engineering for working with the OpenAI API, and I believe that the key principles learned can also be applied to normal ChatGPT usage.
For those who are not familiar with programming but have access to ChatGPT, or for those who have no access to ChatGPT, I have provided a link to a GitHub page where you can follow along with the video. The link is:
https://github.com/xyang2013/prompt-engineering-examples/blob/master/I2-chatgpt.md
If you’re interested in learning more about this topic, I highly recommend checking out the short course offered by Deeplearning.AI. The link is:
https://www.deeplearning.ai/short-courses/chatgpt-prompt-engineering-for-developers/
I hope you found this video informative and helpful. If you’re interested in AI and want to stay updated with my latest content, consider subscribing to my channel. Thanks for watching!
Timeline:
00:00:00: Background
00:02:07: The GitHub page for non-programmers and people without ChatGPT access
00:03:53: Principle 1: Be specific and clear
00:05:00: Tactic 1: Use delimiters
00:07:42: Prompt injection
00:10:05: Tactic 2: Ask for structured output
00:13:24: Tactic 3: Check whether conditions are satisfied
00:15:46: Tactic 4: Few-shot prompting
00:18:18: Principle 2: Give the model time to think
00:19:30: Tactic 1: Specify the steps to complete a task
00:24:41: Tactic 2: Self-checking
00:30:00: Model limitations – Hallucinations
00:35:29: End notes
