2025 in Reflection: Waypoints and Milestones

It’s that time of year: Time for an annual reflection. The thought that looms largest in my mind is, “This is my last year in my 30s.” Next year, I’ll be turning 40, which seems surreal. Am I … middle-aged? When did that happen? For the first time in my life, I feel like I’m counting down instead of counting up …

Last year, I titled my update “In Reflection” because I wasn’t sure of my long-term chances surviving leukemia. This year, I’ve added the subtitle “Waypoints and Milestones” to reflect both the permission I’ve given myself to look ahead and the milestones I will achieve—turning 40 being chief among them.1

Goals for 2026

  • Professional Development: In a lot of ways, I feel I’ve reached diminishing returns from reading books. I still read, but I’ve found myself reaching out more to peers in my industry for their insights and advice. For 2026, I intend to work closer with 2-3 of my industry peers to compare notes on how we’re evolving our engineering organizations in the world of AI.
  • Deliver three to four more apps with my kids while they’re young. These projects have gotten more interest and engagement from my kids than any remodeling, woodworking, or car repair projects I’ve dragged them into. The goal is to spend time with family.
  • Writing: My goal is still to publish 6-10 posts a year. I setup a micro-blog and categories on this site to focus my approach to publishing. In 2025, I read On Writing Well, which I will try and apply to my posts. My favorite suggestions are: earn the reader’s attention, one big idea and succinctness.

Family

I was being a dad. That was my job. And I didn’t have to ask myself “What am I doing here?” … I understood my purpose. –Bob Odenkirk.

The most important accomplishment of 2025, in addition to reaching my two-year in remission milestone, was that my health strengthened enough that I could spend more time with my kids outside. We took several excursions, but I would say the day trip to the New York State Fair was the best of all. Nearly 10 hours on my feet up and down the boulevard with unlimited rides and prizes for all of the kids.

Walking that long was unthinkable even just six months prior; to be able to be present for my kids was the most wonderful family accomplishment of all.

On a more sober note, one of my grandmothers passed away in 2025. This was challenging; the number of people who have known me my entire life seems to shrink with each passing decade. Combined with my own brush with mortality, it’s the strongest reminder I have that family is first, code is ephemeral.

Self

Health is a crown that the healthy wear on their heads, but only the sick can see it. –Al-Shafi’i

The biggest personal accomplishment in 2025 is two years in remission; I claim this—surviving—as my defining achievement. With my long-term prospects improving daily, I’ve been able to focus more on what I want to accomplish with the time I have remaining. One idea has crystalized in my mind about my self-identity as a builder and creator. If this blog is any testament, I really enjoy making things. This clarity of purpose helps me target the types of roles and projects I will pursue in the future.

2025 was also a year of acceleration via AI. It has enabled me to get further than ever before on projects and work, but has also enabled my worse tendency to chase down ideas at the expense of others. I have registered far too many domain names this year in pursuit of future ideas. This year, I really leaned into establishing my Wiglaf Software brand, under which I published Just Cat Mazes a small no-ads maze game I built with my six-year-old. I’m looking forward to building more apps with my kids next year as well.

One thing I left behind in 2025 was OMT Studios. Not only did I discover I don’t enjoy creating YouTube “content,” but a band (https://www.the-harbours.com) reached out about purchasing the domain. I was all too happy to let it go, and I wish them success in their journey.

Engineering

AI is fundamentally changing the way software is written; I plan to shift down and relearn what it takes to ship software and what the expectations are from engineering leaders in an age when code can be continuously generated. I initially experienced AI in 2023-2024 as an error-prone, incoherent code generator. But in the past year, it has proven invaluable to me in expanding my reach and getting more work done.

Learning what does and doesn’t work with AI is critical to my integrity of leading with credibility as an engineering manager. I’m also learning how to mix traditional application development and LLM non-deterministic outputs. What I had previously dismissed as “just” a statistical generation machine can be made far more powerful by stacking prompt guidelines, context, and structured outputs. More on this next year as I find out more myself!

Significant Revisions

  • Dec 30th, 2025 Originally published on https://www.jsrowe.com with uid 725A3ECF-1D75-4FBE-8193-B80B82A56996
  • Dec 22nd, 2025 Draft Created

Footnotes

  1. knock on wood. ↩