January 13th, 2025: Greetings from this side of the near year from hill country in New Braunfels!
Over the past couple of weeks, I’ve let myself do whatever felt right. Since becoming a parent and deciding to spend a lot of time with my daughter, my work hours have become more structured and I’ve more or less “worked” a bit less. In doing this, I’ve become more efficient while prioritizing certain kinds of work, like book writing, above others.
A casualty of this approach has been my tinkering time.
In Good Work, I call these “supporting activities,” ones that give you energy, are fun, and grease the creative wheels that enable good work. But they can be a distraction if they absorb all of your time. This is why in the process of writing my book, I paused my podcast and slowed this newsletter down a bit. Without that, I likely wouldn’t have finished the book.
Someone who’s productivity-pilled might see this as a step forward. For me, it’s mixed. I loved writing and completing another book but I’ve also lost touch with the more playful and experimental “play” and tinkering time that has led me to accidentally build things that made money and laid the groundwork for more interesting creative work.
But over the last few weeks, on my self-imposed work break, I’ve been doing the opposite of finishing the year strong. There were no last-minute sprints or goals met. I’ve been resting, hanging out with my daughter, nursing a stomach bug, reading for curiosity and enjoyment, and this past week, accidentally learning how to code with AI.
I say accidentally in the sense that I had no plans to do anything like this or even any intentions to learn to code in 2025. I just had time and saw that my friend Nat (who I’ve seen build many apps suspiciously fast this year) launched a course on “Building Your Apps with AI” earlier this week.
I had tried to learn to code a few times
I have a solid foundation of the basics and have hacked quite a few things together in C++, HTML, CSS, and JS over the years but never could get over the initial friction needed to really learn how to build anything more complex.
Recently, I’ve become curious about trying again. Since Anthropic released their sonnet 3.5 model, many people have been raving about how good AI-assisted coding has gotten. A few weeks ago, I played around with Replit to try to translate some materials from Angie’s book into English. After 5-6 hours of tinkering, I learned a bit about how to use Anthropic’s API and did successfully translate one text document from Chinese to English but faced enough roadblocks in making it repeatable that I gave up.
But after diving into Nat’s course, I finally got over the hum. His course made it easy. It just sort of flowed into my brain and everything worked.
The key to this could be simply Cursor, or it could be that Nat is a great teacher (honestly a bit of both). Cursor is impressive though, especially its “composer” option, which responds to your queries, automatically writes code, and helps you deploy it. I mean, look how easy it is to “push” the code to github:
As I dove in, things started snowballing over about 24 hours:
#1 Built a Pomodoro timer in under 30 minutes
This was the test assignment (public link here) and I was sort of shocked how easy it was.
#2 Turned a tool I created in Excel, my “freelance target income calculator” into a web-based calculator
This was pretty cool as it was something I tried to do before (and failed). Here is the calculator embedded on my WordPress site (link here).
#3 I built a blog post-re-purposer using AI of all my writing on pmillerd.com
This is another thing I tried and failed to do with Replit. This time with Cursor, I was able to just download an XML file of all my blog posts and have it automatically build a parser to turn the XML file into something I could send to an LLM. In the past, turning XML data into simple-to-read content was not a simple task.
I would have paid for a simple tool like this in the past. Now I can just whip it up in an hour.
#4 Deployed a static website in 20 minutes
I’ve made several attempts to deploy a simple static site on Github but could never get it to work quite right.
With Cursor, I was able to stand up and publish a landing page for pathlesspublishing.com in under 20 minutes.
This is all pretty awesome
While I did eventually hit some roadblocks when I tried to build more complex software, this tinkering has reawakened a playful curiosity with “work” that has been under-expressed for a while.
It was easy to prioritize this mode while living on 25k a year without a child to care for living in Asia but has been a bit harder with higher costs and more responsibilities in the US over the last three years.
It’s funny. At the beginning of last week, I was feeling a bit stuck, struggling to pull together an annual reflection of the last year.
Instead of finishing that, I played around with Cursor for a week. No regrets.
This has helped me remember that I don’t have standard goals. I’m not trying to build the most impressive and profitable business. I’m simply trying to stay on a path that I enjoy and this kind of intellectual play is a key element for those aims.
Now, my mind is exploding with ideas. I’m not sure what I’ll do with these new powers. But I’m looking forward to having fun with it.
So remember, prioritize your own tinkering time too.
+ By the way, if you have any ideas of things you’d love to see me build either with my books or other things, let me know!
+ You can check out the course here or you could probably just install Cursor and try to learn on your own and watch some YouTube explainers. Nat graciously offered $100 off with code PATHLESS. Feel free to use a link with my affiliate code or not.
#2 Traveling Village in 2026
I’ve written about the traveling village before. In 2024, they did a four-month trip to Vietnam, Thailand, and Japan. In 2026, they are doing it again and will be stopping in Taiwan!
Angie and I are planning to participate in this and be very active hosts in the Taiwan part of this. If you are interested, check it out here.
They are also running a 40-day RV trip in Europe this fall before this one which some people may enjoy.
Happy 2025!
I’ve been having fun and hanging out online since 2015. I’ve somehow figured out how to hack a living doing things like this for more than seven years. Wild.
If you like what you read here, you’ll probably enjoy my books The Pathless Path and Good Work. If you’d like to meet others on “pathless paths”, you can join The Pathless Path Community.
Some affiliate links I endorse: Crowdhealth, an alternative to US health insurance, Function Health, comprehensive blood testing and health screening, and Nat Eliason’s Build Your own AI Apps course.
If you received this and find yourself in a state of outrage thinking, “How the hell is this drivel allowed in my inbox?” I strongly encourage you to unsubscribe below. It is your god given right and there is plenty of good stuff online.
A reminder: I don’t check unsubscribe alerts and never look at my subscriber list. So if you feel like unsubscribing, you can do so below.
Signed up to Nat’s course too (your fault!) and loving it. Fills a gap in my skills that was stopping me from building stuff I want - will def be looking to automate posts from my work that looks brilliant
Great read Paul, I'm really feeling a desire to tinker at the moment so this one landed