I’ve been quiet for a few weeks. And I want to explain myself.
Three months ago(-ish) LinkedIn Learning launched my ‘How to Research and Write Using Generative AI Tools’ course. It went on to be hugely popular, so they asked me to do some more courses. Initially it was five more. Then eight. And I’m not quite sure where we are now.
On Friday, the first three of these courses shuffled onto the platform.
Yes, three courses. In one day. I can’t believe it myself!
They are:
- How to Boost Your Productivity with AI Tools
- Using AI as Your SEO Assistant
- ChatGPT Prompts for Jobseekers
All of them share some handy-dandy prompts that you can use immediately in your role. And I hope you find them super-useful.
I’m already getting some lovely feedback for them – but I’d love to know what you think. All opinions welcome!
I’m still beavering away on the next courses. But in the spirit of this newsletter, I want to share with you how I create my courses. Because I’ve only recently realised the way I create them is by pure experimentation.
Let me explain.
Prompt engineering is a new skill. There are no best practices to lean on. There are no official textbooks to crib from. There are no real experts (I define myself as an explorer rather than an expert). So I have no choice but to play around in my efforts to discover how best to use it. Here are some ways I do that.
Try out other people's approaches
I started by watching lots of YouTube videos and devouring every blog post I could find on the subject. Most of them were pretty awful but there were some gems in there. I tried out lots of prompts to see what worked best. And I was soon able to separate the smart thinkers from the charlatans. I then spent time refining the approaches to improve them further. This was a big part of my early journey.
Hunt for the principles
When I’m trying out approaches and developing my own methods, I look for patterns and learnings. I ask questions and keep my eyes open for the unexpected. I note down what I learn and then use those to develop frameworks. This takes curiosity and observation. And a fair amount of head scratching.
Develop hypotheses and test them
I give myself plenty of opportunities to ask “I wonder if…”. This is a vital part of the process to me. I’ve currently got a few hypotheses that I’m waiting to test. How does your level of politeness impact the quality of responses? Do different languages lead to different qualities of response? Do bad grammar and spelling errors have an impact on your responses? I’m currently trying to work out ways of testing them. (If you speak another language, please get in touch so that we can test the language one together!)
In recent weeks, this has led me to a new analogy for what I do. I now see myself as a jungle explorer who hacks through the foliage hunting for knowledge. When I find it I run back and make a map for the group to make it easier for them to get there.
It may sound stupid or pretentious but it’s helping me understand my own process better.
How do you learn new stuff? I’d love to hear your approach.
