The Internet's Two Biggest Complaints Are Gardening and Job Hunting. The Answer Isn't Another App.

If you asked a room of founders to name the most-complained-about problems on the internet right now, you'd hear AI agents, crypto wallets, whatever shipped on Product Hunt this morning. The actual answer, from a database of more than 400,000 complaints we read across Reddit, Hacker News and the open web, is gardening and job hunting.

"Home gardeners fighting weather, pests and soil" and "job seekers stuck in endless rejection" are the two largest complaint clusters we track, each over 2,200 distinct people and growing every week. Bigger than any developer-tool complaint. Bigger than the AI pains everyone is racing to solve. The two loudest problems on the internet are dying tomatoes and unanswered job applications.

Plenty of founders garden, for what it's worth. What almost none of them do is think of gardening as a market. That's the tell. The bubble isn't about where you spend your weekends, it's about what you let yourself take seriously as a business, and a whole class of enormous, everyday problems never makes that cut because it doesn't look like a startup. So the same narrow strip of the economy gets a thousand tools and the rest gets forum threads.

Look closely at these two and you get the same message from opposite directions: the answer is almost never another app.

Gardening: the answer is physical

When we map the solutions people actually need against what exists, they aren't software. They're physical. IoT enclosures that deploy shade and misting over a raised bed when the heat spikes. Soil kits that pair NPK sensors with a custom-blended correction mix instead of just telling you your dirt is bad. Heat-stress and pest biologicals that help a plant survive a black plastic pot in August. And here's the interesting part: most of these already exist, but only at commercial scale, for greenhouses and industrial farms. Nobody has brought them down to the balcony and the backyard. The gap isn't "a gardening app." It's taking a solution that already works for a farm and making it work, and cost, for one anxious person with six tomato plants.

Job hunting: build the boring side of the desk

This is the harder, humbler case, and honestly the one I'm least sure anyone can crack. Nobody could accuse founders of ignoring it. There are a thousand CV builders, job boards, AI cover-letter tools, application trackers. That's the obvious layer, and it's completely saturated. So we looked at the less obvious ideas, the ones a clever founder reaches for next: AI agents that chase down why you got rejected, referral marketplaces that route around the resume black hole, hostile-persona interview simulators, paid micro-project "bridges" that give juniors real experience. Every one of them, when we checked, is already crowded or occupied. Careersim runs the hostile-interviewer sim. Parker Dewey runs the paid micro-projects. Levels.fyi owns the transparency data.

Notice what every one of those tools has in common: they all aim at the candidate. Not because the candidate is where the problem is, but because the candidate is who downloads an app, the easy consumer sell. Meanwhile the actual cause sits on the other side of the table: employers who never reply, an ATS that swallows applications whole, a hiring process with no incentive to close the loop. The person suffering isn't the person who can fix it.

Which is exactly where the real opening is, and where almost nobody builds. When our engine looked past the candidate apps, it kept surfacing employer-side ideas: B2B systems that sit inside the company and force the loop closed. Offer-workflow tools that hard-link a "yes" to an actual letter. Layers that make ghosting visible and accountable. Plumbing between the hiring manager and HR. Unsexy, B2B, hard to sell to a busy HR team, and partly guarded by the big HCM incumbents, every reason a founder skips it for a prettier consumer app that changes nothing. So the honest answer isn't that job hunting can't be solved with software. It's that the software that would actually solve it lives on the boring side of the desk, and boring is what the bubble refuses to build.

Look one vertical over

Put the two together and the lesson is the same. The obvious, glamorous, consumer-app version of a problem is almost always the crowded one, and almost always the one that doesn't work. The real answer usually wears a costume the bubble doesn't respect: a physical product, a boring B2B tool, a marketplace, a service. Building for other builders, or reflexively reaching for another slick consumer app, is a tiny, crowded corner of a very large economy.

So the practical move, the thing I'd actually do: look up from the screen, at the verticals you already touch and quietly dismiss. The trade your cousin works in. The hobby you file under "not a business." The industry whose software you complained about last week. The best opportunities are sitting in the parts of the world the software bubble finds beneath it, and the winning answer is often a physical product, a marketplace, or a service, not the reflexive SaaS. Building is cheap now. The scarce thing is the willingness to build somewhere unfashionable.

We read the complaints so you can see past your own window. The feed's free if you want to look. But the lesson costs nothing: the biggest problems aren't hiding. They're just wearing work clothes.