Tell Your Computer What to Do, and Watch It Deliver Says Browser Use Founder

“Tell your computer what to do, and it gets it done.” – this was the vision of both the founders of Browser Use, when a couple of months back this startup was just an ideal weekend experiment to see if LLMs could navigate the web like humans.

The founder of Browser Use Gregor Zunis and Magnus Müller, during their many conversations and discussion with their discord community of 5000 people quickly realized a few core challenges in browser automation.

Web scraping automation faces issues like website changes, bot detection, and rate limits. Using tools like Selenium or Puppeteer helps adapt to site updates, while rotating proxies and CAPTCHA solvers avoid detection. Simulating human behavior with delays and headless browsers reduces bot risk. For rate limits and parsing errors, backoff strategies and robust selectors can improve reliability. Automating login with headless browsers removes manual credential input, making automation more efficient and resilient.

Therefore they built Browser Use, to help users connect their AI agents to the browser effortlessly. This tool simplifies browser automation, providing a powerful and user-friendly interface that allows AI agents to interact with websites.

Since then it has grabbed a lot of attention from developers of AI agents that can autonomously complete tasks for users as they can complete tasks autonomously on behalf of users.

Since its launch, Browser Use has experienced rapid growth, with over 50,000 GitHub stars, making it one of the fastest-growing open-source AI projects. It has gained significant traction, with more than 15,000 developers actively using and contributing to its tools. The project has been applied in diverse real-world scenarios, such as login automation, data extraction, QA testing, and CRM integrations, showcasing its versatility and widespread adoption.

Browser Use, which has become well-known in recent months, is part of Y Combinator’s 2025 winter batch. Manus, the growing AI “agent” from Chinese startup Butterfly Effect has incorporated Browser Use to execute many tasks using their AI agent. This garnered popularity for the platform even more.

The web is the world’s largest source of unstructured information by a long way, and browser use helps make it more “readable,” which is especially useful for AI agents. The web is also constantly being updated, making it the number one source of fresh information.

According to an interview with Tech Crunch, one of the founders said, “A lot of agents rely on vision-based systems and try to navigate websites through screenshots, and in [the] process, things break,” he said. “We convert [websites] into something agents can understand. This approach means we can run the same tasks again and again at a cheaper cost.”

Within four months of this massive development, the startup has raised a substantial $17 million on seed funding, led by Felicis’ Astasia Myers with participation from Paul Graham, A Capital, and Nexus Venture Partners.

The founders firmly believe that this round of funding will let them build the infrastructure that would enable AI to interact with the web as seamlessly as humans do. 

They are aiming to convert website interfaces into structured text that LLMs can process inevitably, which means providing faster, more efficient automation by directly interacting with web elements like buttons, drop-downs, and input fields, unlike vision-based automation. This method ensures precise UI interactions and avoids the errors associated with pixel-based navigation, making the system more reliable and consistent across different tasks and layouts.

Astasia Myers of Felicis emphasized the necessity of AI agents that will revolutionize software interaction. The Felicis team held the conviction that Browser Use offers the most straightforward approach to integrate these AI agents with the browser via a robust yet user-friendly browser automation interface, and this funding solidified their position.

Holger Mueller of Constellation Research Inc. said, “Browser Use is doing for AI what companies like Google have done for people with disabilities, making websites more accessible,” he said. “With today’s funding, its tool is going to get a lot more publicity and a lot more use, and then we’ll see how well it works at scale. Don’t be surprised if browser makers look to get in on this and introduce their own, robot-ready functions to aid AI web exploration.”

The founders of Browser Use are definitely aware of the changing web landscape, where AI driven agents are navigating their ways through difficult tasks online. This automated workflow will soon outnumber human interactions on the web, where Browser Use plans to facilitate this transition smoothly.