The Person Closest to the Problem

Anton Osika is the co-founder and CEO of Lovable, the Stockholm-based platform that lets people build apps and websites by describing them in plain language. Less than 2 years after launch, the company is valued at $13 billion, and apps built on it get close to a billion visits a month. Before Lovable, Osika trained as a physicist and created GPT Engineer, an open-source code-generation project that became the seed of the company. He started Lovable because he saw too many ideas, his own and those of people around him, die because software was so hard to build. These signals are from Anton Osika's interview with Marina Mogilko on Silicon Valley Girl.

1. Past the Vibes. Osika is dropping the phrase "vibe coding." In his words, "vibe works for demo, but we're way past that. You don't build your company on vibes." He points to founders now running real businesses on Lovable, including Sabrine, whose startup Plink helps people check whether they can trust someone they're dating and has made tens of thousands of dollars in revenue. Software used to be hard to build and a business used to carry a lot of overhead, and he says a serious platform now removes much of both.
2. Eleven Years In. Lovable's customer data shows most people on the platform have more than 11 years of professional experience. They spent those years close to a problem that frustrated them and lacked the skills to fix it. Osika calls this one of the biggest openings the new tools create: the person nearest the problem can now build the solution. Younger builders show up too, and he credits them with seeing things with fresh eyes.
3. Manual First. Before automating anything, Osika says to do the work by hand. Then break it into pieces you can test one at a time, checking that each one reproduces what you used to do manually. Builders who over-engineer or aim too high from day one often end up close, with one small part that fails or a solution that was wrong from the start. "You want to iterate towards perfection."
4. Service Plus Product. If Osika were starting a business today, he would build something tied to the real world: a product combined with a service. The idea doesn't have to be new. He suggests finding people who haven't adopted an existing solution, in areas like groceries or home care, and making it exciting and personal enough that they switch. That takes what he calls a "hustler mentality."
5. The Dog Portrait Lesson. Mogilko asked how a site that turns pet photos into paintings makes $300,000 a month when ChatGPT can do the same with one prompt. Osika's answer: the founder, Damian, doesn't stop at the image. He mails customers a physical portrait and handled the website, payments and marketing himself. Osika says entrepreneurship usually means combining new capabilities into a service nobody has built yet, and the deciding factors are how hard you work and whether customers care about you.
6. Ten Customers, One Payment. Osika's fastest route to understanding customers is to sit in front of 10 of them, describe what you plan to build, and ask about their pain. Then go to 10 more and try to get one to pay. Lovable's product team still tries to talk to 10 people before building a feature. On pricing, he warns against making the product cheap or free, since only payment proves someone cares; once you have 10 paying fans, make them feel special so they spread the word.
7. Build Your Own Agent. Asked what every knowledge worker should build, Osika says: an agent for yourself. Building it from scratch, or having AI build it, teaches you how agents work, because you can ask it how it was implemented. His rule for agent versus app is that anything holding data is an app, and you add a visual interface when you need to see that data. Lovable's new agent integrations let any app plug into an agent harness, whether that's Lovable itself or Claude chat, so he sees the two merging.
8. Inbox to CRM in Minutes. On camera, Osika built Mogilko a lead form that sorts inbound brand deals by AI-scored relevance and drafts replies. He then had Lovable run an end-to-end test and a deep security scan before publishing. The test sent real emails, and the AI agreed to a sponsorship on its own, which Mogilko said she'd rather approve first. Osika calls Lovable the connective layer for a business, with data storage, email, payments and Slack alerts in one place.
9. Same Cost, Hosting Included. Mogilko asked whether running a business on Lovable costs more than using a Claude project. Osika says the building cost is about the same, and hosting and analytics come included. Users get free daily credits, and the default paid plan is $25 a month.
10. The Moving Line. The point where you need a developer keeps moving, and Osika says it depends on how dedicated and curious the builder is. Something at Facebook's scale, serving hundreds of thousands of users at once worldwide, is still outside the line. Most of the software he uses day to day is inside it, and he says Mogilko's study-abroad booking platform, which took several developers 12 years ago, could be built on Lovable. Builders who know their customers ship more polished products than teams that hand work off every few weeks.
11. Creativity Gets Scarcer. "The ideas are becoming more important," Osika says, and creativity is the biggest differentiator as execution gets easier. He thinks everyone has creativity and that the way to improve it is to build, test ideas and learn from the feedback. He still struggles to pick which idea to pursue. His test is to find the bottleneck: whether you need to talk to more people, market better, or fix bugs that hurt retention.
12. More Companies, Less Overhead. Osika doesn't spend much time on competition from ChatGPT or Claude. He says Lovable does "creation of businesses for our customers," and that agent-built software is in its very early days. Looking 5 to 10 years out, he expects money and careers to matter less to many people, while building a business stays one of the best ways to grow and create value for others. Lovable's goal is to cut the cost of running a business so low that many more companies can exist.

Watch the full video at https://www.youtube.com/watch?v=KEAYPqQG2-M.