Conversation With the CEO of a $200M Company

A record of a long conversation I had with the CEO of a ~$200M company — part interview, part networking, part the kind of talk that makes you articulate things you'd only half-figured-out. We got introduced through a mutual connection I'd met on LinkedIn. I've cleaned up the transcript and organized it by topic, but kept it as the back-and-forth it actually was. Names and companies are removed. Me is me; CEO is him.

TL;DR

• The work I'm proudest of: proving a dev tool lives on SEO, then building a template-page engine — developers share deploy templates, every template mints an SEO page, so page count scales with the community, not my hours. Three traffic peaks (a community template, a popular open-source install guide, a cluster of CMS queries) took us to #3 in Taiwan / page two in the US, one post to ~10K pageviews/month, and ARR up nearly tenfold.
• The channel I was wrong about: personal branding. For a niche developer market it's bottom-of-funnel, not top — and the influencer play is ~daily posting for a year with no early signal. Wrong trade.
• My own startup, and why it closed: targeted early-stage startups + traditional industry with GTM automation; died in a co-founder loop of proving who's right instead of focusing on the user.
• Career: high-school self-study → dropped out → interned at a cloud company and a telco → learned by doing. Bar for going back to school is a top program. Big believer in risk-taking (Antifragile: tear the muscle to grow it).
• Why this company: a rare Taiwan company with a real US foothold; its "Shopify for the underdogs" wedge is arming the rebels against a $20B+ incumbent.
• AI + GTM: it collapses content and outreach cost and raises the bar for juniors — your edge is leveraging it to scale output 10–100×.

How we connected

Me: Finally meeting you in person. A mutual connection introduced us — he connected with me on LinkedIn a while back, and I thought he was pretty cool. His co-founder had a background in databases; that CEO was the kind of person who'd seriously read the history of mathematics, so I was curious about the people in that world. He told me about you and about the company, and honestly he said a lot about the environment here that I found genuinely delightful.

CEO: So your previous startup wrapped up, and now you're looking at the market?

Me: Right. Though I'd rather reframe myself as go-to-market. I tend to look at the whole funnel — top, middle, bottom — figure out where the gap is and what we can fill. From experience, it's often not purely marketing. Sometimes sales and marketing have to push together, and there's a human trust issue in the middle. Converting a LinkedIn post into a sale isn't just a content problem — it can go all the way down to product.

CEO: You've done a startup yourself, so you understand how much GTM matters. But we're relatively more mature now, so we have a specific functional need.

The developer platform: biggest takeaways

CEO: Tell me about your time at the developer platform.

Me: My biggest takeaway is a much deeper sense of ownership. You're in charge of everything, you have to think about everything, you have to learn everything. On the business-operations side, I started out not even understanding what a C corp or S corp was, and had to bring the company up from there. On the product side — do you use a North Star metric this early? Linear or Notion? Which tools actually earn their place? Then the product itself: how many people do you interview, who, and how do you turn their pain into a product? How do you cut the MVP — V1, V2, V3? And then actually doing cold-email outreach to your ICP. I learned how to do the job more efficiently because I had no choice but to.

CEO: You know Clay?

Me: I do. Clay aggregates people-data from a bunch of sources into one place. We're a bit like that on our new data product — we pulled multiple platforms' data into a single hub so you can access it under one billing. We tried Clay but didn't use it much; our industry is niche enough that broad search doesn't help, and we had a good data source of our own.

CEO: Clay's expensive, as I remember.

Me: It is — but if it works, the GTM tools that are expensive are usually expensive for a reason.

The SEO engine — the work I'm proudest of

CEO: I care a lot about iteration. You have a hypothesis, you design an experiment to test it, and usually the hypothesis is wrong. The result points you toward the right one. You can't run too many hypotheses at once — there isn't time. So, framed that way: what do you think your biggest contribution was? Contribution isn't only good results; a bad result with a learning is a contribution too.

Me: My biggest contribution was proving that a developer-tool company genuinely needs SEO. I somehow connected with the head of DevRel at another deployment company and asked how they grew — his answer was basically "we greedily wrote articles." Around the same time, a PM friend told me bluntly that our SEO was terrible. So I actually went and studied it. I'd been writing Go; our other engineers were strong, so I moved into go-to-market. We had a little traction — lots of events in Taiwan, around 7,000 developers in our Discord across China, the US and Taiwan — but SEO never took off and revenue wasn't climbing.

CEO: So what was the hypothesis?

Me: That we needed a large volume of templated pages. A developer platform where you can deploy any open-source software — people Google "how to deploy X," "how to install X." Each of those is a great SEO opportunity. So I thought: what if we build a template platform where developers share their templates, and every shared template generates a page — and each page is an SEO page? Get the SEO right and we capture all that traffic.

It worked, and it worked because developers wanted to share templates anyway. We just built the platform; they generated the pages, which generated the traffic. At launch we already had a lot of traffic — ranked #3 in Taiwan for the core keyword, and around page two in the US.

CEO: Page-two US traffic might be bigger than Taiwan's.

Me: Right. The keyword was the product's own name plus how-to, comparison, what-is — "how to deploy X on our platform," "the most X in 2026," and so on.

CEO: So you had roughly three iterations to get to the big win? Walk me through them.

Me: Three peaks — a couple of small wins, then the large one. The first was a community template that already had discussion on Reddit; we wrote the educational version and posted it to our blog, X and Reddit. The second was an install guide for a popular open-source tool — famous in our niche, so I was surprised at the search volume. I wrote the cleanest guide I could, seeded it to Hacker News and Reddit, and US traffic started arriving. The third was a cluster of CMS queries.

CEO: And the numbers?

Me: One single post did around 10,000 pageviews in 30 days. Page views went from double digits to hundreds to thousands, and in the US broke five figures. Over that stretch we grew ARR nearly tenfold. Signup conversion moved from roughly 1–2% up to 3–4%.

CEO: Why did conversion improve?

Me: Mostly the traffic arriving — a bigger, better-qualified base. We also changed some landing-page copy, but that was secondary. The traffic was the win.

Personal branding — the channel I was wrong about

CEO: Did you try things that didn't become peaks — that you thought would?

Me: Personal branding. I really wanted to build one — I kept struggling with it. There are tools that string all your accounts together so you can post everywhere from one place; I badly wanted to use them.

CEO: Why did you want a personal brand? Because you thought you could become an influencer and drive traffic to your own product?

Me: Honestly, I just felt my life was a bit different, so filming it and cutting some vlogs seemed worth it.

CEO: That makes sense, but it's 100× the work you imagine. To actually drive traffic you'd post daily for a year. The days that "work" don't work — you might need 200 days before you see anything. Every blogger's story is like that. For us it's a bad trade. And especially in a niche market, an influencer play fits worst, because the audience is too small.

Me: Right — for a niche it's really bottom-of-funnel, not top. It helps people who already know you decide to trust you.

CEO: Exactly. A CEO or senior leader doing it is bottom-funnel impact — people already know you and see it. For a niche market the target is just too small. Horizontal audiences are smaller than developer audiences in that sense.

The startup I did next — and why it closed

CEO: After that you started something with someone else?

Me: Yeah — wanted to test where my own limits were.

CEO: Any breakout there? What was your B2B ICP?

Me: We were pretty naive about it. We wanted to target early-stage startups — a lot of them need GTM but can't afford to hire for it. Our product was meant to automate the tedious parts of GTM.

CEO: First, small startups. Second — a lot of traditional industry has this need too: few people, but real money. Their problem is nobody there knows how to use the tools, and the tools are expensive.

Me: Right, they won't manage it themselves.

CEO: Did you consider a done-for-you / outsourced model for traditional industry?

Me: I did — but that turns you into a studio, an agency, which isn't the product we set out to build. And it doesn't scale: you have to keep hiring, so it becomes a body shop.

CEO: But with automation, one person can do much more — maybe it scales after all. Especially with AI.

Me: Fair. A friend of mine runs a reseller-type company with lots of sales, and his point was: you can scale revenue by adding people, but the margin stays the same, so you don't actually make more. That's the trap.

CEO: True for sales. But marketing can scale.

Me: I have some doubt on that, but I get what you mean. At the time we thought about partnering with GTM studios and consulting firms to reach their clients.

CEO: So why did you close it?

Me: Honestly, working with my co-founder wasn't happy. Direction wasn't aligned. When we hit a problem, we'd argue — and arguing is fine, that's necessary — but we couldn't turn the debate into an output: a next hypothesis, an experiment, a way to validate it. We got stuck in a loop of trying to prove who was right instead of focusing on the user. That was the real issue, and a big learning. We parted ways.

Dropping out

CEO: Tell me why you dropped out. You were studying in the US and left halfway?

Me: I didn't study for long before dropping out. The honest reason is I didn't know what I was doing. I did high-school self-study in Taiwan, then applied to US universities.

CEO: Then where'd you go and what did you do?

Me: I went and interned — first at a cloud company, then a large telco. I'd studied abroad about a month, came back not long after. Part of it was I couldn't find work there; and the first internship I got into partly through my dad's connection. I knew nothing, so I just showed up early, stayed till six, and saw how much I could learn — and it turned out to be a huge amount. My mentor there — a PhD student doing ML — taught me a lot: who to go to, who to learn from, how to read documentation, how to write a WBS, a POC. So I thought: I've learned this much — do I really need to go back and learn everything school teaches? Is what school teaches actually what I want to learn? I already had a pretty clear sense of what I wanted and what I was missing, so I kept exploring my own way. After that I moved to the telco — same path: learn until you hit a wall, figure out what you're missing, keep going.

CEO: What was the telco role?

Me: Some innovation/new-business unit — tied to a C-level person, more about turning their planning ideas into something built.

Twenty-one, and why risk-taking matters

CEO: How old are you now?

Me: Twenty-one.

CEO: Really young. You can be job-hunting at 21 — that's great. So why do you keep reviewing whether to go back to college?

Me: I do keep asking myself when should I go to college. I could leave and go back. But my bar for myself is a top program — somewhere with people who really feel like mine. To get there I have to do something spectacular first. I want to put myself in the best environment I can, learn from and brainstorm with the best. It's like basketball: play only with people at your level and you improve a little; play with people much better and you improve a lot. Environment matters. Talking with you actually got me fired up about it.

CEO: A lot of people your age want to go into a big company — a US master's, then a FAANG. There are plenty of success stories for that. But I think there are other paths, and I feel like I'm walking one, so I feel special. Your self-taught background is an advantage. Chinese culture constrains people here — you can't lose at the starting line, you can't fail once.

Me: Right — Chinese culture is you can't fail even once.

CEO: But you learn the most from failure. Failure is the mother of success, no? Chinese culture just can't accept it — one bad exam and you're done. So it's very hard to take a risk, and very hard to build something big.

Me: I really do think risk-taking matters. I read about a third of Antifragile — Taleb's cynical, but he's right that you tear the muscle first to grow it.

CEO: Exactly — you tear it, take in protein, and it grows back bigger. Same with a startup: you have to eat some shit and fail a few times, or you'll never know which road is which.

Me: You can listen to how others succeeded, but success paths all differ; failure is where the real, transferable learning is. Still — hearing about it and failing yourself are two different things.

CEO: Completely. However much you read, it's a textbook until you actually do it. And you'll have your own way of failing, because a startup is just too hard, too many dimensions. Even people who succeeded won't necessarily succeed at what you're doing.

Why this company

Me: First reason I like it — I saw your board on an accelerator's site and used to go there and talk to people. I saw your page on LinkedIn and thought it was cool, though the space is niche enough that I didn't fully get what you did at first. But what really drew me: Taiwan has very few companies with a real foothold in the US — not just "attacking" the US market, but actually standing there. That distinction matters to me, because my own career goal is to build in the US. If you have a genuine US foothold, joining means I can learn how to win the US market for real.

Arming the rebels

Me: Then I researched the industry you sell into. Your positioning is basically "Shopify for the underdogs," and that's an interesting pitch — because Shopify's early strategy was arming the rebels. So who are your rebels? I knew there had to be a giant in this market; the mid-size players who haven't been swallowed are your rebels.

CEO: Exactly — the giant is a $20B+ incumbent that's taken almost all the top players in the space. The remaining mid-size companies don't have a good modern system. We give them a SaaS platform so they get big-company IT competitiveness — that's the core of the pitch.

Me: That's a great wedge — a clear positioning and a clear opponent. Lots to do on GTM there.

AI, GTM, and entry-level work

CEO: On AI — what impact do you think it has on GTM and software companies?

Me: Huge, because it collapses the cost of content generation and outreach. Whether it's SEO or B2B sales, work that used to need many people writing articles and emails, AI can now do.

CEO: The most direct hit is SDRs and junior marketing/content people. We're seeing AI agents do first-touch outreach, and even the first round of HR interviews. We demoed a tool yesterday that reads a person's LinkedIn, summarizes who they are, writes a customized email, and auto-replies. There's even AI that does the first screening call by voice.

Me: Right — that voice AI can hold a real back-and-forth, ask questions, react to your answers.

CEO: So does AI replace juniors? I think it raises the bar. It speeds up the cycle: your AI reads my email, my AI writes yours. It's a productivity jump — you still need people, but at a higher level. A junior who doesn't learn to use these tools, design prompts and frameworks, gets replaced fast — because a boss realizes a few-hundred-dollars-a-month tool can replace two or three junior SDRs.

Me: Completely agree. That's exactly why I keep trying new AI tools, writing scripts, wiring up APIs. If you can't leverage these tools to scale your output 10× or 100×, you don't have an edge.

CEO: And you want to take career risk instead of the traditional finish-college-then-big-company path — that mindset is good. At an early- or growth-stage company we really need people with a founder mindset who go find answers and test hypotheses themselves.

Me: The most valuable thing from my past roles is exactly that — you hit the wall, you eat it, and you know how to walk next time so you don't die.

CEO: Good talk. I'm clear on your background and what you're chasing. We'll sync internally this week and have HR update you on next steps.

Me: Thank you for the time — I learned a lot. Looking forward to it.