A 19-Year-Old With No Resume Raises $6.2 Million

Zero Lines of Work Experience, $6.2 Million, Age 19

Here’s a résumé for you. Education: dropped out of high school in Kazakhstan. Work history: none. Age: 19. Any ordinary recruiter would toss this profile out at the screening stage. But the person behind this résumé, Arlan Rakhmetzhanov, is now the founder of Nozomio, a San Francisco startup running on $6.2 million in funding—roughly ₩8,500,000,000 (~$6.2M). He started coding at 15, cold-messaged Y Combinator¹ alumni on LinkedIn, and landed his first angel check at 17.

What did investors see in a teenager with not a single line of corporate experience to justify handing over that money? The answer is that the yardstick investors and companies use to evaluate people has changed. Instead of a résumé listing where you belonged, the new standard looks at the public record of what you’ve actually built. Let’s look at how this shift happened, and what shape it’s taking in Korea’s hiring market.

Looking at the fragment, I don’t see any Hangul characters, and the numbers all match the Korean source (July 31st, 19, $1.6 million, 20, 14-15, 18, 30, 2022, 24, 2025, 110%, 2018, 2015, 2022, 29, 30, 21). The glossary terms don’t appear in this fragment. Structure (paragraphs, bold, em dashes) is preserved.

One issue: “amounted to ‘premature optimization’” — checking the Korean “warned that it was ‘premature optimization’” — this is fine, matches “warned.”

The translation reads naturally and accurately. No surgical edits needed.

Founders Under 20 Are Raising Seed Rounds

This isn’t just Rakhmetzanov’s story. TechCrunch ran a piece on July 31st digging into the world of founders under 20, and the profiles all read strikingly similar. Pranjali Awasthi dropped out of high school to launch an AI startup, enrolled at Georgia Tech, dropped out again, and built Slashy, an email-focused take on Cursor. She’s 19 now, and recently revealed she’s quietly building yet another company on the side. Aidan Guo, 20, raised roughly $1.6 million for an AI desktop-assistant startup.

Awasthi’s own account of the shift is striking. Around age 14-15, meeting investors meant getting hit with “why do you even want to start a company?” as the opening question — but now that she’s past 18, that question has simply vanished. The numbers point the same way. The median age of Y Combinator founders fell from 30 in 2022 to 24 in the 2025 batch. One analysis found that the number of accepted founders aged 18-22 rose 110% in a single year.

What’s interesting is that this runs against Y Combinator’s own long-held philosophy. Founder Paul Graham, in a 2018 interview, named the late 20s as the ideal founding age and warned that starting too young amounted to “premature optimization.” And indeed, from 2015 through 2022, the average age of participants hovered right around 29. Wasn’t Silicon Valley supposed to love young dropout founders in the first place? True — but there were strings attached. You needed a technical co-founder to pair with, or at least a line of Big Tech experience on your résumé.

Ashley Smith, a partner at early-stage investor Vermilion, explains what’s moved in to fill the space where that condition used to sit. These days, she says, investors look at a founder’s GitHub activity, history of open-source contributions, communities they’ve personally built and grown, and their fluency with the newest AI tools. Young developers contribute to open source and tinker with new tools while learning to build software — and, as Smith notes, they simply have more time for that than someone holding down a full-time job with a mortgage to pay. Sure enough, founders under 30 make up a meaningful share of Smith’s portfolio, several of them not even 21 yet.

The upshot: résumés now count for less in investment screening, while public track records get checked first.

The Résumé Was Always a Proxy Metric

Why did this substitution become possible? We have to start with what a résumé actually is.

A résumé and a school pedigree aren’t direct evidence of ability — they’re proxy metrics. Economist Michael Spence’s 1973 signaling theory² explains this: a degree’s value lies less in the knowledge actually learned than in the signal that its holder was able to clear the gate. The reason companies and investors have leaned on proxy metrics is simple. Observing ability directly was too expensive. You can’t know until you’ve actually put someone to work, so instead people borrowed someone else’s filter — Stanford’s, or Google’s.

But over the past few years, both of those costs have fallen sharply, together.

First, the cost of building. Thanks to AI coding tools, teenagers can now build products alone and put them on the market. That means you can produce a real track record on your own, without ever joining a Big Tech company. Rakhmetzhanov’s Nozomio started exactly that way — beginning as an AI coding agent that reads an entire codebase, and now building an API index that helps AI agents find and use outside software.

Second, the cost of observation. Whatever gets built this way ends up publicly logged anyway — in GitHub commit histories³, open-source contribution records, app metrics, community size. Evaluators no longer need to borrow someone else’s filter; they can look at the raw data themselves. When the cost of directly verifying ability drops, the value of a proxy metric drops with it. The résumé’s shrinking weight is simply the consequence.

To be fair, the idea that open-source activity functions as a hiring signal isn’t new. Economists Josh Lerner and Jean Tirole already identified career-signaling as one of the key motivations behind programmers contributing to open source for free, in a 2002 paper. What’s changed is the scale. What used to be a workaround for a handful of developers is now standard due-diligence material for investors.

Still, this shift has a darker side. As verification has sped up, there’s less room to wait out early failure. Smith says the market no longer grants the tolerance that early-stage companies used to get — the assumption that if you iterate enough, you’ll eventually find product-market fit. Even though that growth curve is an obvious outlier, she says, everyone’s out looking for the next Cursor. Timothy Chen, an investor at Essence, makes a similar observation: startups used to worry about the incumbent giants; now they worry about the same-age startup in the next seat over. The glossy launch-video arms race that didn’t exist 3 years ago is evidence of that. In the words of Guo, all of 20 years old: the fear of failure is always in your head, because everything can go wrong at once, and the moment you slip up, people swarm in to tear you apart. He adds that this kind of hostile social ecosystem simply didn’t exist back when Zuckerberg was building Facebook.

A generation selected by public records keeps getting evaluated by those same records afterward. Rakhmetzhanov’s line — “build a company as valuable as Google, or end up on the street” — isn’t the bravado of a 19-year-old. It’s a sentence that precisely internalizes a market where the outlier has become the baseline.

Korea: Mass Hiring Fades, Job Records Take Its Place

What about Korea? The grammar of hiring has been moving in the same direction for quite some time now.

Let’s start with the shift from large-scale, seasonal mass hiring to rolling, job-specific recruitment. Hyundai Motor scrapped its regular mass hiring in 2019, and LG and SK followed. Of Korea’s four largest conglomerates, only Samsung still runs a regular hiring cycle. Samsung is the company credited with introducing open recruitment to Korea in the first place, back in 1957 — so the firm that opened the door to mass hiring is now the last one holding it shut. In a 2025 survey by the Korea Enterprises Federation (KEF), 70.8% of companies with 100 or more employees said they now hire exclusively on a rolling basis, and 85.8% said they recruit whenever a need arises, with no fixed hiring season at all. A separate survey of Korea’s 500 largest firms by the Federation of Korean Industries (FKI) found that rolling recruitment now accounts for 63.5% of hiring — up 5 percentage points in just a year — and that 28.1% of last year’s college-graduate new hires were so-called “seasoned newcomers” (junggo sinip): graduates who already had prior work experience before landing their “first” job.

The disappearance of mass hiring matters to today’s story for a specific reason: mass hiring was the institutional perfection of the proxy metric. To filter tens of thousands of applicants at once, you need standardized signals — school pedigree, GPA, language test scores. What’s replacing it, job-specific hiring, asks a different question: what have you actually built? This shift is furthest along in software development. Coding tests have replaced document screening, and it’s already standard practice for a GitHub link and portfolio to get opened before the cover letter does.

But there’s a distinctly Korean wrinkle here too: whenever a signal changes, an industry springs up to sell that signal. Coding bootcamp ads promise a future where a single GitHub link is enough to get hired, and job-seeker forums swap tips on “planting grass” (jandi simgi) — filling in the commit-activity graph with green squares to fake a track record of daily coding. It’s the same old list of résumé specs job seekers used to pad, just renamed as a list of things you’ve supposedly built. This is Goodhart’s Law⁴ in action: the moment a metric becomes a target, it stops being a good metric.

How far this competition over signals can go is best illustrated by a case from the US. Roy Lee, founder of Cluely, built a tool called Interview Coder that secretly helped software engineers cheat their way through technical interviews — and got suspended from Columbia University for it. But instead of sinking him, that history became a selling point: Cluely went on to raise a reported $20 million cumulatively, in a round that included Andreessen Horowitz. In other words, money from investors armed with the new tools of verification flowed to the person who built a device for defeating those very tools. Verification mechanisms and the tools built to beat them keep provoking each other into evolving together.

Oswarld’s Lens

I find this shift both welcome and worrying.

The welcome part comes from a data perspective. When I teach data science, I tell my students something over and over: a snapshot and a time series are completely different kinds of evidence. A resume is a snapshot — you can dress it up right before you submit it. A commit log is a time series — three years of consistency can’t be faked overnight. What investors trust isn’t the GitHub platform itself, but the fact that a record accumulated over a long stretch of time is hard to fabricate after the fact. I think this direction is fundamentally right.

The worry comes from my consulting experience. Working on go-to-market strategy, I watched companies evaluate talent up close, and I saw a pattern: organizations that only measure what’s measurable end up undervaluing capabilities that resist measurement, relative to their real worth. Things like collaboration, judgment, and accountability don’t show up in a commit log. And as grass-planting shows, any observable signal eventually gets commoditized. When that happens, what an investor checks isn’t how many commit squares got filled in, but what problem you worked on, why, and in what order.

So here’s what I tell my students: don’t try to fill in commit squares — grab hold of a problem you actually want to solve. One repository that reveals what problems this person cares about is worth more than ten repositories stuffed with 100 commits each. And this isn’t homework just for developers. The speed at which “company name” is losing power as a proxy signal is accelerating across every profession, not just tech. Building a long, steady record of output that shows what you did without needing a company name to vouch for it — that’s the preparation that fits today’s market.

Closing

Teenagers with zero lines of work experience raising millions of dollars isn’t really a story about individual success—it’s a signal that the criteria we use to evaluate people have changed. As AI tools lower the cost of actually building something, and public track records lower the cost of verifying skill, the basis of evaluation is shifting from where you belong to what you’ve made. The trend in Korea away from regular mass hiring cycles toward job-specific recruitment is part of the same shift. But the generation selected through public track records keeps getting evaluated by those records afterward—and an industry that manufactures and sells new signals on people’s behalf has already emerged.

I’d suggest trying something this week. Write a three-line self-introduction without mentioning your company name or title. If you can’t fill it in, what you need to build isn’t credentials—it’s a track record.

Which side are you on, Reader? Whether you’re a hiring manager who opened an applicant’s actual work before their resume, or an applicant who’s been judged by what you made—tell me about that moment in the comments. If enough stories come in, I’ll put together a picture of how this plays out in Korea in the next issue.

---

💬 Share in the comments any experience where you evaluated—or were evaluated—based on something you made, instead of a resume. I’ll factor it into the next issue. 📨 If you have a colleague preparing for a job search or career move, pass this along to them.

References & Further Reading

Primary sources

• TechCrunch, “Build in public, fail in public: what it’s like to be a founder under 20 right now”, 7/31/2026. Link ··· This is where today’s piece starts. It carries the actual voices of Rakhmetzhanov, Awasthi, and their investors.
• Inc., “Y Combinator Bet It All on AI. Now Founders Are Wondering If They Should Bet on YC”, 5/2026. Link ··· The source for the median age of Y Combinator’s 2025 batch being 24 (versus 30 in 2022).
• Euclid Ventures, “No Country for Old Founders”, 5/2026. Link ··· This piece lays out Y Combinator’s shift in data, from Paul Graham’s warning against “premature optimization” to the 110% jump in accepted founders aged 18-22.
• Michael Spence, “Job Market Signaling”, The Quarterly Journal of Economics, 1973. ··· The original source for the idea that a degree is a signal, not knowledge. This research won the 2001 Nobel Prize in Economics.
• Josh Lerner & Jean Tirole, “Some Simple Economics of Open Source”, The Journal of Industrial Economics, 2002. ··· A paper identifying “career signaling” as a motivation behind open-source contributions. It’s essentially a 20-years-early preview of today’s story.

Background

• Korea Employers Federation, “2025 New Hiring Survey”, 2/2025. Link ··· The source for the finding that “70.8% of companies use rolling recruitment only.” Based on a survey of 500 companies with 100+ employees.
• Federation of Korean Industries, “H1 2025 New Hiring Plans Survey of Large Corporations”, 3/2025 / “H2 2025 New Hiring Plans Survey of Major Companies”, 9/2025. Link ··· The source for the 63.5% rolling-recruitment figure and the 28.1% figure for “experienced new hires.”
• Seoul Economic Daily, “SK Hynix declares ‘no-credential hiring’… Samsung did it 31 years ago”, 6/2026. Link ··· Notes that among Korea’s four biggest conglomerates, only Samsung still runs a fixed-cycle mass hiring program.
• TechCrunch, “Cluely, a startup that helps ‘cheat on everything,’ raises $15M from a16z”, 6/2025. Link ··· Traces the growth of the credential-forgery industry, from interview coders to Cluely.

Recommended past issues

• Issue 170: Six Claims of a Perfect AI Score, Only Two Officially Graded
• Issue 164: Why Freelance Developers Moonlight as Delivery Riders

---

The author is Oswarld (Kwangseob Ahn). Current roles: Adjunct Professor at Sejong University, Strategy Consultant at INLEVEL9. Career, research, books, and recent work are kept current on the About page. Latest · July 2026: HEMA-2: A Consolidation-Aware Tri-Memory Architecture with Multi-Channel Scheduling for Lifelong Conversational AI.

📝 Glossary

Footnotes

1. Y Combinator: A leading American startup accelerator founded in 2005. Airbnb, Dropbox, and other well-known companies passed through it, and acceptance alone functions as a strong seal of approval for early-stage startups. ↩

2. Signaling Theory: An economic theory describing how someone conveys their otherwise invisible abilities to a counterpart who lacks information, through indirect signals. Degrees, certifications, and brand names are classic examples of signals. ↩

3. Commit Log: A record generated every time a developer saves a change to code. It stacks up chronologically — when something changed, what changed, and why — functioning as a kind of work diary for that person. ↩

4. Goodhart’s Law: The principle that once a measure becomes a target, it ceases to be a good measure. This happens because people start optimizing their behavior to score well on the metric, undermining the metric’s original purpose. ↩