I started last Monday. Today it's been exactly two weeks.
Here's the funny part. I've spent 15 years in QA — manual testing, then automation, then test development, then test management. I've held just about every title this field has. And on day one, at a new desk, the feeling was as basic as it gets: I have no idea what I'm doing.
Why here
The way it happened was mostly luck. After my last company laid me off, I didn't really look for work. I spent the time on dev.to, writing my 36 Stratagems series — 30 down, 6 to go. That stretch was honestly pretty good.
Then an old colleague called. He was hiring testers and asked if I knew anyone. I thought about it and said: how about me.
One conversation with the CTO, one HR interview, and then it was Monday.
Why I said yes is simple. It's an AI agent startup, a year or two old, sitting right on the hottest thing in the industry — and in a vertical, not another general-purpose wrapper. That was enough on its own. Poking at agents and models is what I do for fun anyway.
Two things I noticed in two weeks
The first is process. The product → requirements → dev → test → ship pipeline here is genuinely different from what I'm used to. I'll be honest: the first week I thought it was a mess, and I'm still adjusting. When you're used to one kind of order, a different kind just looks wrong. Usually it isn't wrong. It's different.
The second is pace, and this one's good. My last company ran on instant replies. Requirements changed, and the PM was expected to answer on the spot. Here it feels slower, in the best way. Testers get more time and more room to make their own calls. You're allowed to think something through instead of getting dragged along by it. Two weeks in, this is my favorite thing about the place.
Do QA long enough and you learn something: slow is fast, and fast is slow. Time you steal up front gets paid back with interest, in rework and firefighting. Half a day spent on the edge cases, the weird inputs, the what-ifs — that half day usually buys back a week. So when I say this place feels slow, I don't mean nobody's in a hurry. I mean the pace goes where it counts.
Understand first, judge second
Fifteen years buys you fast pattern recognition. You look at something and the problem jumps out at you. It also buys you a bad habit: when something doesn't match how you'd do it, the first instinct is to doubt it instead of figuring out why it's that way.
So my rule for these two weeks: understand first, judge second. A beginner's mind sounds easy. In practice it's an argument with yourself.
The coincidence that makes me laugh a little less
I wrote a character in the 36 Stratagems series. His name is Mark.
Mark was the reliable one — years of experience, all of it in his head. His company slowly extracted that experience into a skill. Then they laid him off.
I wrote that one from the outside, as the author, feeling pretty pleased with myself.
In that story, the Skill scored 96.8% diagnostic accuracy across 312 historical failure scenarios. Then number 313 showed up. The 450ms retry window was a compatibility shim I'd written five years earlier for RabbitMQ, and it got applied to a system that had been running Kafka for years. I'd even left a note in the migration docs: 450ms matches RabbitMQ GC window. Do not reuse outside this context. At the time, nobody understood it. They thought it was a stale config comment. Not until 4 AM, when the CTO went looking for it.
Here's the line I closed that post with:
The AI didn't fail because it was wrong. It failed because it was right about yesterday — and yesterday wasn't running anymore.
Then I joined this company. And on the walls and along the corridors, there are two signs:
Your experience is waiting to be forged into an Agent.
Great employees get the work done. Great Agents keep getting it done.
(In Chinese, since that's what they actually say: 你的经验，正在等待被炼成 Agent。／优秀员工完成工作，优秀 Agent 持续完成工作！)
Me: 15 years in QA, most of the titles in this field checked off at least once. By the standard of those two sentences, I'm not an employee. I'm raw material, still in the wrapper.
And look at that second one again. Great employees get the work done. Great Agents keep getting it done. So I do the doing, and it does the keeping. I raise it, it replaces me. Structurally speaking, the logic holds up.
Back when I was writing Mark's story, readers laughed. These days I walk under those signs and I laugh a little quieter. Part of me wants to read that closing line out loud, right there in the hallway: it wasn't wrong, it was just right about yesterday.
Though honestly, being forged into an Agent doesn't bother me much. Tinkering with agents and models is what I do for fun. I'm genuinely curious what 15 years of me would come out looking like.
One request, though: take your time. I'd like to finish those last 6 first.
Fresh eyes
One more detail. After that Mark post went up, a reader left a comment pointing out that my timeline didn't add up — the same parameter described as a one-year-old thing in one place and a five-year-old thing in another. I'd gone through that story more than 30 times. Never caught it.
I replied in the comments: you read it more carefully than I wrote it.
That's the job, honestly. AI can run 312 cases. Case 313 still needs a human. And humans miss things too, which is exactly why you always want a pair of fresh eyes on it. Fifteen years in, that's still what I get paid for.
What's actually worth extracting
Writing that Mark story is what made it click. The valuable part of experience was never the number 450ms. It was the context for why it's 450 and not 300 — context that came out of a 3 AM postmortem and never made it into a single doc.
So if these 15 years do get forged into something, I'd want more than the conclusions in there. I'd want the judgment behind them, the part that's hard to trace back to a source. Which is also why I wanted to work at an AI agent company. Not because I'm worried about being replaced. Because I want to find out, hands on, how this is supposed to be done.
An apology I owe
30 of the 36 Stratagems are done. 6 are left. They're on hold while I get up to speed at the new job. So to my friends on dev.to: sorry. I'm not quitting the series. I just want to learn this business properly first, and then sit down and finish the last 6 the way they deserve. I haven't forgotten what I owe you.
What's next
The story I wrote ended on something nobody ever asked: validation only works when the person who wrote it is still around to re-run it.
That question is part of why I'm here. The AI agent space doesn't lack smart models. What it lacks is people watching to see whether the thing still works. I've done that job for 15 years. Different track, same job.
I also just like this company's agent platform. It's the kind of thing I'd dig into on my own time. So the plan is boring and simple: learn the business, ship what I'm supposed to ship, then write the last 6.
Long road. Slow and steady. It'll go fast.
P.S. English isn't my first language. I use AI to polish the writing and smooth out the rough edges. Thanks for reading. ☕ Buy me a coffee
P.S.S. The old series paperback is now available. Buy it here: Amazon - AI, Ego & Regret
Top comments (22)
"Time you steal up front gets paid back with interest, in rework and firefighting."
I really like this sentence, it applies on all parts of building a product not just testing!
Congratulations on the new job! I also have no idea what I am doing and I feel like a bit of an imposter every time I start a new job. 🫣
Hahaha 😂 We’re all the same. A new place brings a fresh start. Especially for an introvert like me, asking my new colleagues that very first question is the hardest part. But it gets easier after that.
Congratulations on the new job, my friend. 😄
But I have to say, after reading the Mark story, joining a company whose wall literally says “Your experience is waiting to be forged into an Agent” is either excellent career planning or the beginning of Stratagem #37.
Please keep an eye on the documentation. If you ever find a file called xulingfeng-final-v2.agent, don’t open it. 😂
Jokes aside, I think you touched the most important part here: experience isn’t the 450ms value. It’s knowing why 450ms existed, what assumptions made it correct, and when those assumptions stopped being true.
We are getting very good at extracting answers from experienced people. Extracting the boundaries around those answers is much harder.
And yes — finish the last six before they forge you. Some of us are still waiting. 😄
Congrats again. This sounds like exactly the kind of place where your QA background and your obsession with breaking AI systems can cause the right kind of trouble.
Hahaha🤣, love this Stratagem #37 warning! I’ll keep a close lookout for that agent file. You hit the nail on the head: the numbers are trivial, the assumptions and boundaries are the real treasure. Rest assured, the final six stratagems will be published before they turn me into an agent 😂
“I have no idea what I’m doing.” made my day. 😂 All of us, even with years of experience, feel like that sometimes. 😄
Haha glad this line resonates! No matter how many years of QA experience you have, every new codebase and team makes you feel that way at first 🤣
Those wall signs next to the Mark story are a wild coincidence. Good luck at the new place and with the last six stratagems! 😊
Thanks a lot! It really is such an eerie coincidence. I’ll keep pushing forward with both the new job and the remaining six stratagems😆
Great text, Lingfeng! Oh yes! After working at startups and then moving to a big corporation, I've definitely learned to appreciate the slower pace too! 😄
Thank you! I totally get that. A calmer pace really helps with deep work. 👏
Hahaha, two weeks in and you're already walking under signs that could have come straight out of your own story. 😄. Oh yeah! My Sensei is back.
You gave @technogamerz's thought book "fast is slow, and slow is fast" back in August, with the Tao Te Ching recommendation attached. Now you've landed somewhere that actually works that way, and I think that's a really good sign for a new job.
@merbayerp is right that the boundaries are the most difficult part. In Mark's story, the boundary was actually written down. "Do not reuse outside this context" was sitting right there in the migration docs. It still failed, because anyone without the context read it as a stale config comment. So the problem wasn't that the boundary never got written down. It was that a warning looks like noise to anyone who doesn't know why it's there.
That's exactly why the wall signs don't worry me on your behalf. An Agent forged from 15 years of you (Veteran plus Sensei) would inherit every note you ever wrote, and it still wouldn't know which ones are load-bearing. Your "understand first, judge second" rule is that exact discipline: treat everything as possibly load-bearing until you know why it's there.
Health IT in England has a formal version of this. Under DCB0129, each clinical safety control is recorded in a hazard log against the hazard it mitigates, and the log is kept live as the system changes. It's basically your 450ms note with teeth, meaning this one bites because I have lived experience.
And there's an Igbo proverb that fits your first two weeks perfectly: Onye ajụjụ anaghị efu ụzọ. The one who asks questions never loses their way.
No apology needed for the last six. They'll be better for the wait. Take your time, Sensei!👊
Hahaha 😂 It’s almost like I’m writing prophecies. I made up fictional stories, and then reality served me the exact same experience 😂. I really love that proverb, it makes so much sense. At work, never feel ashamed to ask when you don’t understand something. Many colleagues here are younger than me. Whenever I’m confused about business logic or operations, I just ask directly. Grab a few meals together, and we quickly become friends, hahaha.🤣
Hahaha, then please give Mark a very happy ending in the last six, Sensei, just in case reality is still reading. 😄
And sharing meals is the proverb in action. The context that never makes it into the docs usually gets explained over food. Trust me, I have been in one of those hangouts!👊
Hahaha, exactly. Many things get sorted out over meals. If one meal doesn’t work, have two. Lunch buddies are the most efficient team combo in a company.🤣
This was such a relatable read, especially the part about going from being the person who already knows how things work to suddenly feeling like “the new guy” again.
I think there’s something really valuable in that experience. When you’ve spent 15 years in QA, you naturally build a mental library of patterns — you’ve seen flaky tests, weird regressions, production bugs, bad requirements, “it works on my machine,” and probably every variation of “we’ll fix it later.” :D That experience is incredibly useful, but I also liked the way you described having to step back and actually observe instead of relying completely on that experience.
The AI angle makes this even more interesting. We’re all getting used to tools that can generate answers, write tests, explain code, or even suggest what might be wrong. But knowing what to ask, recognizing when an answer doesn’t quite make sense, and understanding the context behind the answer still feels like a very human part of the job.
The distinction between knowing an answer and understanding why that answer is correct really stood out to me. Anyone (or anything) can give you a number, a test result, or a recommendation. The harder part is knowing whether that result actually makes sense in the context of the system you’re testing.
And honestly, I think being “the new guy” again can be a surprisingly good thing. It forces you to question assumptions that have become invisible over the years. Sometimes the person who has been around for a long time knows how something works, while the new person is the one who asks, “Wait, but why do we do it this way?” Those questions can be uncomfortable, but they’re often where the interesting problems are hiding.
Also loved the honesty in this post. It doesn’t read like “15 years of experience and now I have everything figured out.” It feels much more like, “I’ve got experience, but I’m still learning how to apply it in a completely different environment.” That mindset is probably more valuable than the experience itself.
And the idea of having to earn your way through the Stratagems was a fun touch. 🙃 It made the whole experience feel less like a checklist and more like an actual journey of figuring out how this new world works.
Really enjoyed reading this!!!!!!!!
Hahaha 😂 Maybe in a few months, my refined agent will be the one reading and replying to these comments. But I will definitely forbid that. After all, reading and replying to comments is my favorite part right now. The new company is pretty great, and the boss is generous. It’s worth sticking around.🎉
Glad to see you're finding you sea legs 😂 adapting to a new company is always a tricky one, especially trying to fit in, instead of making them fit to you. You're right on the 'it's probably just different' and the understanding part is where you make the difference. Some things you did before are arguably better, so it's up to you to identify what it'd replace before pitching it. Who knows, maybe they structure their CI differently, but you know that pattern costs more over time, because it tests unnecessary things and it's better to split it? The small things bring lasting value and lasting value is exactly why you were hired.
It's odd how some fear being replaced by AI, whereas some of us know it's inevitable and just want to make sure when it does, it wont make the same mistakes we did. That mental shift in the current AI space is exactly what keeps the industry growing at a stable pace. There has to be those who push back to make those who push forward prove their worth and it's essentially red-teaming 😂
You'll do great though, cuz it sounds to me like the company fits your ethos.
Then I’ll keep my fingers crossed for your new job. Wish you all the best. Praise what deserves praise, and complain about what needs complaining. Life is like a play; we have to act it well. The whole world is a big stage.🤣
Bro really said “slow and steady” and then wrote a whole essay 😁
🤣忽然有感悟，就忽然发了一篇，想发就发，再不发感觉就要被遗忘了
Some comments may only be visible to logged-in visitors. Sign in to view all comments.