Hello hello.
Welcome to a new issue,
This time we’ll look at using AI to speed up even more and how the sleep experiment went.
Let’s go.
Main quest 🛡
Not ignoring that AI in general can help in speeding up our learning process but also not using as a clutch to replace that process completely.
We have to be conscious of how and when we’re using it.
Why am I telling you this?
Because I’ve already have my fair share of seeing “influencers” online claiming that you can become an expert in X topic if you use their Chad GeePeeTee prompts.
Which sounds like those people a few years ago that claimed you could read the top 5 books on Amazon about a subject and become an expert at it.
Yeah right. 🙄
So before I go on a long rant on how that is not only plain wrong but downright offensive to folks with real expertise…
I want to share with you one practical way to supercharge your learning while not outsourcing your brain to an AI.
Use the expert interview.
If you recall from a previous issue, I told you about how I found out the top performers in area and ask them questions to elucidate the path to proficiency.
But not just any questions, increasingly better questions to get much better answers.
Which is something that nowadays people are realizing more and more.
Ask normal questions, get normal answers. Ask great questions, get life-changing insights.
Now, you can skip the back and forth or the trial and error by using an AI to play the role of the expert and give you a glimpse into that which you don’t know.
Here’s an example prompt you can use:
You’re an expert in [chosen topic] with years of experience in [example of accomplished goal]. Your goal is to answer the following questions to the best of your ability drawing from all the years of experience in the field. Here are the questions:
And here are some of those expert questions to ask the LLM.
What advice would you give to someone starting out in [chosen topic], and what are some common mistakes to avoid?
Who are the most unorthodox [performers (like runners, artists, developers] and trainers?
Who trained or influenced you?
What are the biggest mistakes and myths that you see others doing?
What are the biggest time wasters in [particular field]?
What are your favorite instructional resources on the subject?
If people would have to teach themselves about this subject, what would you suggest they use?
What are some of the key mindset shifts you’ve seen helped others and how would you suggest to develop them?
These are some of the questions that you can get started with. There are more but I don’t want to overload you with too many questions that you end up not getting any answers.
Notice that the questions are set up in a way that will give you insights but will not give you the full answer to the problem.
Because if the brain doesn’t work to solve a problem but instead get the answers in a silver platter, it will not learn and will not be able to recall them at a later time when it’s necessary.
Related to what I mentioned in this tweet.
So, go ahead and give it a try.
(Bonus points if you use a “mixture of experts” type of LLM like mixtral-8x7b-instruct or dbrx-instruct)
Let me know how it works for you and what area/topic you used the prompt & questions for.
Experiments 🧪
Following up from last week, this week I had 2 days where I woke up between 6 and 6:15 am without needing an alarm clock. In only used the alarm for 6:10 am once.
And I have been tracking the time I go to bed daily to make sure I don’t go over the 10:30 pm mark.
Now I know that I can do a 11:30 pm to 7 am type of schedule. But sleep science tell us is that the night hours are much better to get into deep restorative sleep.
In that case shifting the wake up time 1 hour is better for many more reasons than just “sleeping earlier”.
Of course this is highly dependent on chronotypes. So make sure you know which one is yours and shift hours around accordingly.
To consider 🤔
If you could have the opportunity to get really good at something, like going from zero to hero in an area, would you choose an specific one or a few related ones?
That’s it for this week.
Thanks for your time.
Now, if you’ll excuse me, there’s a few friends calling my name.
Catch you next time.
Juan.