What I got wrong about studying in China
Notes from Ganzhou, and one conference that rearranged how I think about the gap.
I came to Jiangxi University of Science and Technology expecting to learn computer science. I did learn computer science. That turned out to be the least interesting thing I learned.
What I expected
I expected the technology to be ahead. Everyone says this, and it's true in a narrow sense and misleading in a broad one.
The code isn't magic. The algorithms are the same algorithms. A distributed systems lecture here covers what a distributed systems lecture covers anywhere, because the material doesn't care what country you're in.
What's ahead isn't the knowledge. It's the distance between having an idea and holding the object.
The thing that actually shocked me
I could sketch something, and have a physical version of it, quickly. Not because of any single company — because the entire supply chain sits close together. Components, tooling, assembly, logistics, all within a range that makes iteration a normal Tuesday instead of a quarterly project.
That density is the advantage. Not intelligence, not work ethic, not some cultural story people like to tell. Proximity, compounded over decades of deliberate industrial policy.
It changed what I think Africa needs to copy. Not the factories — you can't copy your way to that, and trying is how you end up with an industrial park nobody uses. What's copyable is the density itself: clusters where the people who make a thing are near the people who assemble it and near the people who ship it, so the cost of trying something drops far enough that many people try many things.
The part nobody warns you about
The language barrier is not what you think. It isn't conversation — you can get by, people are patient, translation apps work.
It's that daily life runs inside apps that assume you read Chinese and have a local identity. Payments, deliveries, the campus systems, appointments. Being locked out of the software layer is a more complete isolation than not speaking the language, because the software isn't patient and doesn't gesture.
The first months were administratively humbling in a way I had not budgeted for. Things that should take ten minutes take an afternoon, and none of the afternoon is spent on anything you could call learning.
And I'm often the only Nigerian in the room. That's mostly fine and occasionally strange. It does mean that when people form an impression of a place from one person, I'm the sample size, which is a weight you carry quietly.
The conference
I attended the International Conference on Artificial Intelligence and Big Data, organised in part by the China Computer Federation. Sessions on generative AI, large language models, visual SLAM, the framing of the arc from narrow AI through generative systems toward something more general.
I went in expecting to feel behind. I came out with a different feeling.
The frontier work is genuinely hard and genuinely impressive, and nobody in that room had a secret. The confusion about where this goes is evenly distributed. People disagreed with each other, publicly, about fundamentals.
That was the useful part. Not any single talk — the discovery that the distance between a serious researcher and a well-read practitioner is smaller than I'd assumed. The gap that matters isn't knowing. It's being close enough to the work that you can try things and find out.
Which is the same lesson as the supply chain, arriving from a different direction.
What I'm taking back
Not the technology. The technology is available to anyone with a connection and some patience.
The thing worth carrying home is the assumption that the distance between idea and object should be short — and that when it's long, that's a fixable condition and not a fact of life. Most of what makes building hard in Lagos or Nairobi is not a shortage of talent. It's that every step takes three weeks because the next step is far away.
Close the distance and the talent shows up on its own. It's already there.