Never again,apparently

I taught myself. Design, then HTML, then CSS, then enough PHP to be dangerous.

By third year I was freelancing. One person doing the whole stack, because there was nobody else to hand any of it to.

In 2004 I stopped freelancing and went on staff at ITB — VCCI. Several e-commerce portals over three years.

I was hired to design them, and I could build them too, so I ended up in both conversations — what the thing should look like, and what it was actually for. The second one turned out to be the one I couldn't leave alone.

That was the first time I noticed it. The design part was never the part I cared about most.

I chased it.

iBOSS Technologies ran two things at once: accounting software, and web services for small companies. I had the web half.

At the end of the first year I sat down and counted what I'd made. Nearly one hundred layout designs. Just me.

I still remember the number, twenty years later. I don't remember much else about that year.

I left my own company to go work for someone else.

Five years at ePi, full-stack design for some of the largest news sites in the country. I played a major role in a news aggregator that went on to take that market here.

Same pattern as VCCI, larger stakes. Full-stack design meant I could argue about how something should work and then go build the front of it, which made the arguments considerably shorter.

The interesting problem was never in the layout. It was in what the product was for.

I moved.

One of the five was the news app I'd helped build at ePi. I'd worked on it from the outside for years. Now I was inside the company that owned it, which was a strange way to arrive.

Six years. I built the team from zero — product designers for the products still finding their shape, and a creative team running the advertising campaigns.

The traffic in Ho Chi Minh City took me about a year to understand. Scooters weaving across every lane, no visible rule, and somehow nobody dies at the intersection. There's an unwritten rhythm holding it together.

It took me longer to notice I'd chosen the same thing at work. On purpose. Twice now.

I asked anyway.

I became Product Manager for the news product — which was partly the org's idea by then, and fair enough. I'd designed for major news sites for a decade, and I'd already been advising that team on interactive story formats.

Three years on reading experience. How people actually move through an article. What makes someone finish it.

The doubter showed up somewhere around here. You had a team. You had a title. What exactly are you doing.

I don't have a good answer to that one. I still don't.

Eleven months on a payment gateway. Ideation, planning, execution, and a great deal of explaining things to stakeholders.

I was not the smartest person in most of those rooms.

That's the part I'd forgotten I liked.

Rebuilt it.

Eleven years earlier I'd built a design team from nothing, which sounds like the harder version and isn't. Starting from zero means nobody is standing there while you work out what you're doing. This time there were two people already in the room, watching me decide.

Then I spent most of my energy on the part I actually came for: getting AI workflows adopted across design and product. Orchestrating agents. Prototyping web and native myself, hands on the tools, for the first time in years.

Building again, and deciding how the building gets done.

Which is what I was doing in 2004, on e-commerce portals nobody remembers, with worse tools and no idea it was a pattern. I have apparently been after the same thing for twenty-two years and I only noticed it while writing this page.

Which brings us to the actual reason any of this happened, and it isn't a strategy. Every one of these was a mess I could see from the outside, and I walked into it anyway.

Same as the mountain. Halfway up Poon Hill I promised myself I would never do that again. I'm currently looking at a ten-day trek.