The Complete Prompt Engineering for AI Bootcamp
Brief
This course broadened my understanding of AI services and how to think strategically and pragmatically with prompts. The instructors covered the full gamut of AI systems used today. I gained hands-on experience with image generation (Midjourney & DALL-E) to chat assistants (GPT-4) to generating entire articles. Even though the video content and walkthroughs were 13.5 hours in total, I spent twice that amount of time experimenting with their examples.
Techniques Learned
Aside from the tacit knowledge imparted by the instructors, there were five principles that applyed to every system. Here they are:
- Giving Direction
- Specifying format
- Providing Examples
- Evaluating Quality (hallucinations can get out of hand)
- Dividing Labor
In all, there were 20 sections in the course. The assignments at the end of the sections were brilliant and forced me to think of ways to induce creativity and accuracy in outputs.
Applying Knowledge
There are so many applications of this knowledge, today.
Thinking Futures
If tokenization is predicated on design thinking, (meaning, research findings based on evidential user needs in context), then there is a lot of room for growth.