Hello Again
June 8, 2026 · blog, ai, design
After 15 years working in software, I decided to take a deep breath and start something new.
Why now?
We are at the beginning of another productivity shift. At the same time, the gap between the vision and the day-to-day reality still feels huge.
On one side, frontier labs are racing toward AGI, and it can feel like many white-collar jobs — especially software engineering in the narrow sense — are only a few months away from being automated. On the other side, a traveler in a foreign country can still get stuck because they cannot comfortably speak the local language.
There is a lot of noise in between: anxiety, FOMO, the fear of falling behind, and a kind of frantic chase that keeps getting louder. I believe the technology matters. I also do not want my life to be driven only by that narrative.
I find myself caring more about simpler things: craftsmanship, curiosity about the world, and real connection between people. These things did not disappear. They just often get pushed down the priority list.
I also realized that I have lived inside the software-and-bits ivory tower for a long time.
In the past, making a change like this was hard. The opportunity cost was high, and there was no clear opening. AI changes the equation. It amplifies what I am already good at, but more importantly, it helps cover my weak spots. It lowers the cost of entering unfamiliar domains and communicating with people outside my usual world.
That opens up a lot of room to explore.
What’s next
After leaving my regular job, I started a one-person company: Flintwright LLC.
The name came from a long back-and-forth with ChatGPT. “Flint” is the stone that makes a spark. “Wright” means a maker or craftsman. Roughly speaking, Flintwright is a craftsman who makes sparks.
That is close to what I want to do: help more people and small teams — myself included — learn, start something new, and achieve more.
This is still a broad goal. Many things are yet to be determined. But it feels like the right reason for me to start my own company, and it is the kind of problem I am interested in pursuing.
I keep thinking about The Goal: A Process of Ongoing Improvement. It is a management novel about a factory that is close to being shut down, and a plant manager who has 90 days to find the real bottleneck and save it.
The most memorable story is a hiking trip with a group of kids. The speed of the group is not determined by the fastest kid. It is determined by the slowest one. The same is true for a factory: system output is not decided by the most impressive local optimization. It is decided by the bottleneck.
I think AI is in a similar place today. The models are already very strong, but the bottleneck in the larger system is not always model capability. It may be product design, workflow integration, trust, cost, training, or the basic question of: how does this fit into what I actually do every day?
Those problems are less glamorous. They are also much more real.
Restarting the blog
Build in public is popular right now. For a one-person company, I think it is especially useful.
The hardest part is not having ideas. The hard part is continuing to move. Writing in public can bring friends, feedback, and a little extra pressure to keep going.
So I am restarting this blog. I may also make more videos over time. I want to keep a steady rhythm and share what I am working on, what I am learning from using AI, and the products and ideas behind Flintwright.
AI will be part of the process, of course. I will use it to write, edit, organize material, review my own thinking, and probably make plenty of mistakes along the way. I want to record that process too — partly as a way to improve my own work, and partly because the rough edges are often where the useful lessons are.