Building Lovable: $10M ARR in 60 days with 15 people

Anton Osika, co-founder and CEO of Lovable, explains how his AI software engineer turns plain-English prompts into working products, and how the company grew to roughly $10M ARR within two months with a 15-person team. The conversation covers a live demo, the origin story from the open-source GPT Engineer project, how the team hires and operates, and his views on which skills matter as AI builds software.

01Key learnings

• Be precise and explicit when prompting an AI builder: say exactly what you expect and which parts are not working, rather than just saying something is broken.
• Use the agent's chat mode to ask how something works or why it isn't doing what you want; this builds understanding while you build.
• Expect to spend a full week taking one real problem from idea to a working product that people use; Anton suggests that alone puts you in the top 1% of AI tool users.
• As building gets cheaper, the scarce skills shift toward figuring out what to build and judging whether the result is good, meaning discovery and product taste.
• Hire generalists with as many skill sets as possible: architecture, product taste, user conversations, and enough technical depth to understand constraints.
• Use paid work trials, from a day up to a full week, to evaluate candidates in practice, and screen for obsession with what they've built and an appetite for intense, high-urgency work.
• Start products in an AI-native environment and add AI where it solves specific problems; retrofitting AI into an existing product end-to-end, as Anton's earlier startup found, is hard.
• Keep product planning lightweight: weekly planning, a shared idea board, and a rough roadmap that is expected to change month to month, with engineering leading solution choices.

Summary and takeaways written by PM Atlas; quotes are short excerpts. © the original author.