My first full-time job handed me two bugs and one lesson I’m still unpacking twelve years later.
September 2014, Kuala Lumpur. I’d just converted from part-time to full-time, the youngest and least experienced person on the team. Everyone around me had more years. My manager had more than ten in the industry. I assumed, the way juniors do, that years translate into answers. Two problems broke that assumption in the same season.
The search that returned nothing
The first problem looked small. Users misspelled their search queries, and the search returned no results. Type the product name perfectly and you’d find it. Miss one letter and you’d get an empty page, as if the thing didn’t exist.
The team discussed it. Then my manager reported to management that there was no solution.
I wasn’t convinced. Something about “no solution” felt wrong for a problem this ordinary. So I proposed building a tool myself, some way to match a wrong spelling to the right product. What I didn’t know, and what nobody on that team knew, was that this problem had been solved for decades. Fuzzy matching, edit distance, n-gram tokenization, entire search engines built around exactly this. I was sketching a wheel without knowing wheels existed.
Here’s the part that stings. I didn’t lose to the problem. I lost to authority. The person with the most years said there was no solution, and I gave up without searching, without investigating, without spending one evening checking whether the world had an answer. His ten years outweighed my doubt, in my own head, where it mattered.
The world did have an answer. I just trusted that someone with more experience had already looked. He hadn’t. Years in the industry tell you how long someone has worked. They tell you nothing about whether they searched.
The black box that wasn’t
The second problem was uglier. A search request took more than thirty seconds to return. Thirty seconds, for a user staring at a spinner.
The official explanation came pre-packaged: the issue is with the service provider. That’s what the external team that built the integration said, so that’s what my manager said. The service was treated as a black box. Seniors didn’t touch it, didn’t question it, and made it clear I shouldn’t either.
This time I didn’t stop. I debugged the code anyway. Then I went around our whole stack and tested the connection directly against the provider. The response came back in a few seconds.
A few seconds. Not thirty.
The provider wasn’t the problem. The problem lived somewhere in our own layers, in code we owned and could fix, hidden behind a label that said “external, don’t look.” The black box wasn’t a technical boundary. It was a social one. Nobody had checked because checking felt like doubting the people above you.
Two lessons, one still unfinished
The first lesson I took: don’t abandon your own idea just because someone with more years dismissed it. Validate it yourself. Experience deserves respect, not surrender. “No solution” is a claim, and claims can be tested, usually in an afternoon.
The second lesson cuts the other way, and this is what makes the pair interesting. My instinct to build a spelling-correction tool from scratch was the right instinct pointed in the wrong direction. The energy was good. The ignorance was the problem. I almost spent months reinventing fuzzy search because I never spent an hour reading about it. Since then I’ve trained myself to search before I build: who hit this problem before me, what did they try, where did their solutions stop.
I’d love to tell you that journey is complete. It isn’t. Twelve years later I still catch myself reaching for “I’ll build my own” before I’ve finished checking what exists. Some weeks I catch it early. Some weeks I’m three days into a prototype first.
There’s a small echo at the end of this story. Years later, on a much bigger platform, I led the introduction of a real search engine, the kind built on everything I didn’t know existed in 2014. Misspellings, synonyms, ranking, all handled. Nobody in that meeting said “no solution.” I made sure of it.
Years of experience are not knowledge. Sometimes they’re just years. The difference is whether you keep checking.