Building Lovable: $10M ARR in 60 days with 15 people | Anton Osika (co-founder and CEO)

Episode Summary

Executive Summary: Anton Osika explains Lovable as a personal AI software engineer that turns plain-English ideas into working products, aiming to make software creation accessible to non-coders and eventually become the "last piece of software." The episode covers Lovable’s explosive growth, demoed product capabilities, team structure, hiring philosophy, prioritization, and how AI is reshaping product-building, skills, and startup execution.

Main Topics: What Lovable is and why it matters (Priority: 5/5): Lovable is positioned as an AI software engineer for anyone: users describe an idea, and the system generates a working product that can be iterated on and launched. Anton frames it as democratizing software creation for the 99% who don't write code. Live product demo and workflow (Priority: 5/5): Lenny and Anton demo building an Airbnb clone from a two-word prompt, then editing UI elements visually and connecting backend components like login and storage. The demo illustrates both speed and the current limits of prompt clarity and iteration. Explosive growth and product-market fit (Priority: 5/5): The conversation emphasizes Lovable’s unprecedented early growth: millions in ARR within weeks, hundreds of thousands of users, and rapid word-of-mouth adoption. Anton credits the growth to strong product quality, not marketing tricks. Technical scaling and getting AI unstuck (Priority: 4/5): Anton discusses the main challenge in AI app generation: models getting stuck in bugs or dead ends. Lovable improved reliability by identifying the highest-value failure points, especially around login, persistence, and payments, and tuning the system around those areas. Team structure, hiring, and generalist skills (Priority: 5/5): Anton argues that the future favors generalists with broad product, design, and engineering understanding. Lovable hires for obsession, speed, taste, and ownership, and uses work trials to evaluate candidates in practice. What changes for builders and product teams (Priority: 4/5): The episode explores how AI shifts value from raw coding toward problem selection, product sense, taste, user understanding, and learning how to guide AI effectively. Anton recommends that product teams become more cross-functional and AI-native. Roadmap and future vision (Priority: 4/5): Lovable aims to enable end-to-end product creation, including deployment, collaboration, analytics, experimentation, and go-to-market support. Anton sees a near future where products can be built, improved, and scaled with much more automation.

Key Arguments: Lovable’s core value is making software creation accessible to the 99% who do not code, turning ideas into working products through natural language. The product’s growth is driven primarily by user delight and word of mouth, not paid acquisition. Fast growth at small team size is possible when the product sits on top of strong foundation models and the team focuses on packaging, reliability, and UX. AI product-building still requires strong human skills: clear communication, curiosity, patience, product taste, and the ability to diagnose what went wrong. Generalists are becoming more valuable because modern product teams need people who can think across design, architecture, engineering, and user needs. The hardest AI product problems are not demos but getting systems to work reliably on important tasks like authentication, data persistence, and payments. Hiring should prioritize obsession, ownership, and evidence of deep care for prior work, not just credentials or narrow technical depth. The future of software building includes not just code generation, but AI-assisted analytics, product iteration, testing, and go-to-market help. Existing developers still matter, but their role shifts toward translating human needs into technical solutions and working as high-context operators around AI tools.

Data Points: Time to first prompt output: 30 seconds - Anton says Lovable’s first prompt can generate a functioning Airbnb-style interface in about half a minute. Monthly active users: 300,000 - Lovable’s user base at the time of the conversation, less than three months after launch. Paying users: 30,000 - A subset of Lovable’s monthly active users are paying subscribers. Initial ARR growth: $4M ARR in the first 4 weeks - Cited as part of Lovable’s breakout growth after launch. Early ARR growth: $10M ARR in the first 2 months - Used to illustrate Lovable’s hypergrowth with a very small team. Team size: 15-18 people - The episode references Lovable operating with 15 people during the fastest growth and later being at 18 people. Open source traction: 50,000+ GitHub stars - Anton says GPT Engineer became a popular open-source showcase of LLMs creating applications. Project creation volume: 15,000 projects per day - Anton says GitHub shut down Lovable-related activity after the product created massive GitHub load. Revenue pace: $1M ARR in a week - Anton says Lovable hit this pace at launch and then continued to grow even faster. Usage scale after launch: 100,000s of users - The product quickly reached large-scale adoption through organic growth and demos. Tool adoption: 17% of newsletter readers use Cursor - Lenny shares a survey result to illustrate how common AI coding tools have become among his audience. Workforce composition: 12 of 18 write code at least part-time - Anton explains that most of the team contributes technically, even if roles are broad.

Pivotal Quotes: "Lovable is your personal AI software engineer." — Anton Osika: His simplest definition of the product at the start of the interview. "We're building the last piece of software." — Anton Osika: Anton’s description of Lovable’s long-term mission and vision for software creation. "The best word for a great product is that it's lovable." — Anton Osika: Explaining the origin of the company name and his product philosophy around "minimum lovable product" and beyond.

Implications: AI is moving software creation from coding-heavy execution toward taste, clarity, and product judgment. Builders who master AI tools, generalist collaboration, and strong problem selection will have a major advantage, while startups may ship far more with far smaller teams.