Leaving Cliwant

Cliwant is an AI startup built for Korea's public procurement market, and after two and a half years there, I'm leaving.

In April 2024, when GPT-4 had come out and AI chatbots were slowly reaching the general public, the idea of building a business on AI felt very new and bold. Cliwant's business model stood out because it was so clear. I could explain it to my mom and she'd get it.

Korea's public procurement market is worth 250 trillion won. Every day, notices for new government projects go up in scattered pieces across a few hundred government websites, and every notice comes with several RFP files (requests for proposal, statements of work, and so on), each around 100 pages long, in HWP, the Korean word processor format. The websites themselves don't show what a project involves or even who qualifies. To find out, you had to download every file and read it yourself. And about 3,000 of these notices pour out every single day.

Naturally, no person can keep up with that. If you're lucky, you spot a notice before it closes. If not, you find it after the deadline and think, I should have gone for that one, and that's the end of it.

Cliwant's three co-founders decided this business would be better with AI attached to it. Right after founding, they raised a seed round from Bon Angels, the fund of Krafton chairman Jang Byung-gyu (Krafton is the Korean game company behind PUBG), and they made revenue from the very first month. The system used AI to screen every requirement in the attached files and recommend the project notices a company could actually bid on.

AI has improved a lot since then, but there's no reason GPT or Claude would ever build a way to download and search every attachment on Korean government websites, and I don't think they'll bother in the future either (unless the government sites change the procurement process itself to be AI-native). Government projects also come with very clear processes and rules. Cliwant writes public-sector proposals that fit them, too: based on a company's internal history and current business, it produces proposals dozens of pages long using the APMP (Association of Proposal Management Professionals) methodology. To do this, three things have to line up: years of public bidding data, the company's own business data, and a bid proposal methodology. Without domain experience and knowledge, you're bound to get lost.

I joined two and a half years ago as the sixth early member. Building a business on AI was tempting, but more than that, there was the moment when Cliwant was selected as an OpenAI partner in Korea just six months after it was founded and showed up all over the media. I found that so surprising that I went looking into what it was all about, and by chance Junho's Brunch post came up in my LinkedIn feed. That's how I came to understand why OpenAI had chosen to work with Cliwant, and then I went on to read every Brunch post he had written. His stories of hustling in Singapore and Hong Kong, with all the ups and downs, and his raw way of telling them, got my heart racing. So I asked him for a coffee chat myself, and luckily I got to join.

To be honest, on my first day, when I walked into a tiny room for eight people in a coworking space, I suddenly got scared and wondered if I'd made the right choice. Once you set foot on a career path, it's hard to switch again.

But since I'd decided to do it, I went in with the mindset of throwing myself into it and doing as much as I could. I asked myself what I'd do if this were a business I had founded, came up with a lot of ideas, and actually carried out a lot of them. Luckily, the company was right there with me in spirit, letting me try as many different things as possible. The team kept growing, and we moved offices three times.

Of course there were hard times, but there were far more times that were fun and exciting. If I had to name something hard, it would be my knees hurting a bit from all the walking on long business trips to the US. When I got back to Korea my knees were sore, so I went to the hospital and did a few rounds of physical therapy. After a few sessions, they don't hurt at all anymore.

Through all of it, I didn't pay any attention to what my friends were doing with their careers. I just lived day by day, focused and absorbed in my own work and whatever was in front of me.

So I honestly never dared to imagine I would leave. Of course people on the team left now and then, but I'd just think, oh, they're leaving, and it never really sank in. Now that I'm the one leaving, I'm a little stunned myself. It almost feels like a dream.

But I truly believe that good people will meet again someday, whatever happens. And while I wasn't perfect in everything I was responsible for, I gave it my all with no regrets, so I don't think I have much left to hold on to.

Leaving something familiar that you've grown attached to always comes with a lot of regret. But life keeps evolving, and good or bad, change keeps coming, so I've decided to think of this as a change that comes from moving forward.

There were actually three reasons I decided to leave Cliwant.

1. America

The first is America. I'm not interested in "global." I'm only interested in the US. But because I have a Korean identity and I live in Korea, I kept finding myself tied to Korean ways of thinking, Korean culture, and Korean ideas. In 2022, I actually sold every last thing I owned in Korea (down to the nail clippers) on Karrot, Korea's secondhand marketplace app, and went on a 10-month trip around the world with my wife. Underneath it was the feeling that anywhere but Korea would do. I wanted a taste of what it's like to live in another country, so I went less as a traveler and more with the mindset of someone moving there. In the end, I came back much more grounded and with a much wider view, but I still didn't manage to move abroad. I wouldn't call it failure. I think I just wasn't ready yet to leave the comfort zone of Korea, where my parents and so much of my support system are.

But having tried once and, in a way, failed, I now want to try again with the mindset of truly leaving Korea behind. Of course, with a second child who is only three weeks old and a first child who is 22 months, emigrating isn't easy. So while I'm in Korea, I plan to spend as much time as I can with people from other countries instead of Koreans, build something like a small foreigners' village in Korea, and get myself used to it. Soon, I plan to make today one of my last posts in Korean and switch all my thinking over to English and a foreigner's mindset.

2. Family

The second is focusing on my family and raising my kids. What I'm worst at is balance. Ever since I was little, when I focus on something, I dig into that one thing very deep and nothing else. The people who liked that about me liked it, and my parents knew that side of me too. Of course I think well of myself when I'm absorbed like that, but in Korea, where the work culture is already intense, I could see myself focusing only on work even more. It was like that from my first job. I justified it by telling myself that Korea has no natural resources, so it has to pour its human resources into everything. Realistically, that isn't completely wrong, but statistically, that's why Korea's happiness index is fairly low for a country this developed.

So in my Brunch posts I thought very seriously about how to live while pursuing happiness, and I read a lot of related books from abroad, plus books on history, psychology, and philosophy. The answer is probably harmony and priorities, and I knew all too well that I'm not good at either. Maybe I wasn't that serious about it.

In the past, maybe I still had a lot to learn and it wasn't yet the time for me to chase balance. But now I'm a father of two and the head of a household, and I wanted to spend more time with my kids before it's too late. When our second was born, I took parental leave and spent some time at the postpartum care center. Every so often I'd come home and spend a night alone with my first child, just the two of us. Playing together without a laptop or work on my mind, almost for the first time in my life, was so sweet and made me so happy. I used to always say you'd have to move to somewhere like Scandinavia to live such a family-centered life. Then I found that just changing my own mindset, more than I'd expected, meant I could try it now.

So right now I'm trying not to pick up work. I want to spend a lot of time with my kids now, while they're at their youngest, because this time only comes once. Since there's daycare and a government-supported childcare helper anyway, I plan to slowly start doing the work I can during those morning and afternoon hours.

3. Freedom

Last, and some people may not understand this, I really hated the life of an office worker. I've only worked at companies in Korea, so I don't know what it's like abroad, but I'd guess it's probably similar. That's why I used to tell my teammates honestly that I thought of Cliwant as my last job at a company. It meant I was going to give Cliwant everything I had. It also means that even now that I'm leaving, I have no plans to move to another company, and no plans right now to found a VC-backed startup either.

Maybe it's even more so because of the current AI trend, but I think that in the future, everyone will have a freelancer's career instead of an office worker's. "Founder" sounds too grand, like you have to build some huge J-curve. What I mean is a way of working where you collaborate project by project and don't belong to any fixed job or group. Whether people want it or not, I keep thinking we'll all end up working in more flexible ways like that.

If the opposite of an office worker is a freelancer, then that seems worth a try too. Of course, in Korea, being a freelancer gets treated as close to being unemployed. I think that's because freelancers get lax and have no coworkers to keep them in line, so I'll probably need a lot more systems in place to keep pushing myself.

When I think about it, I'm not chasing huge wealth or fame. If anything, what I'm chasing is freedom. Of all freedoms, the one I love most is freedom of expression. If I had to pay every time I said what I think, I'd probably pay and keep saying it. That may be what has kept me writing on Brunch for so long, and something like my personal mission.

So if I'm given a freer environment, I'd rather put my energy into expressing my own thoughts with no strings attached, exchanging ideas, building more relationships with the people I want to know, and giving and getting inspiration, than into focusing on making more money on my own. Whether or not it makes money.

My plan for now is to really focus on raising my kids and on my family and build some kind of routine. I don't think I'll do any big, broad community business or activity. I do plan to build a specific group where I can form loose ties with people I like and am interested in. I'm not looking to attach a business model to it. I want to give and get energy and inspiration so I can keep growing too, and just getting to spend time with people like that gives me a lot. My hope is that through that group, I'll find my next step.

Through Bloom recently, I'm not sure what I learned about community as a business, but I learned a lot about the power of people. A company, a school, a family: they're all communities. Inside a community, people who matter to each other clearly lift each other up and help each other grow. I'll find the role I can play and do it.

Finally, thank you so much to CEO Junho Jo, who gave me the chance to be one of Cliwant's early members, to co-founders Guyeol and Seungdo, and to every teammate I worked with. Thank you, from the bottom of my heart, for the unforgettable memories and all the chances to grow.

Originally published in Korean on Brunch, October 30, 2026.