ComposedA startup that turns web browsers into intelligent automation machines has raised $5.6 million in seed funding to eliminate what its founders call “digital grunt work“, affecting hundreds of thousands of data staff.
The round was led by Nat Friedman and Daniel Gross of NFDG, with participation from Menlo Ventures and anthropics Anthology Fund. The San Francisco-based company has attracted users from lots of of corporations including Google, Above, Doordash, Tesla, SalesforceAnd Reddit in only two months since launch.
In contrast to the wave of recent AI browsers from corporations like confusion And OpenAI Requiring users to desert their existing settings, Composite takes a unique approach: It turns whatever browsers professionals already use into an intelligent assistant that predicts and automates repetitive tasks.
“We're not trying to exchange Chrome,” said Composite co-founder and CEO Yang Fan Yun, who was New Zealand's national valedictorian and achieved the very best grade point average in Stanford University's computer science department. “We make the browser you already use again and again more powerful by predicting and automating your work before you even take into consideration asking for it.”
The battle for AI browsers is heating up as OpenAI, Google and startups battle for automation supremacy
The funding comes at a time when the AI ​​agent space is experiencing explosive growth and major players are competing to develop systems that may handle digital tasks autonomously. OpenAI recently released its “ChatGPT agent” skills while Perplexity launched his Comet Browserand Google continues to evolve AI features for Chrome.
But Composite's investors argue that the corporate has found a more practical path to adoption by meeting users where they already work quite than forcing them to learn recent tools.
“Quality and habit formation,” said Matt Kraning, partner at Menlo Ventures, when asked what sets Composite other than the competition. “They meet users where they’re and don’t ask them to vary numerous habits directly, for instance by switching them to a completely recent browser.”
Kraning, who has built billion-dollar category developers, sees particular value within the automations he calls the “messy middle”—tasks that must be done greater than two to 3 times, but lower than ten thousand times. “I exploit it quite a bit to assist research corporations and founders,” he said.
A have a look at Composite's technology: How local AI agents learn your work patterns
Composed Works through a Chrome extension that monitors user behavior to learn individual workflows, then suggests and runs automations on each website. Users activate it with a keyboard shortcut that displays a compact overlay with personalized task suggestions.
The system runs entirely locally on users' devices and addresses privacy concerns which have slowed the adoption of AI in enterprises. “We actually never should ask you to your login information or bank card information,” Yang explained, stating that the system uses login information already stored in users’ browsers.
The company uses multiple AI models under the hood – combining small, fast open source models for specific tasks with larger vision models for complex operations – quite than being tied to a single vendor's ecosystem.
“We use best-in-class models from multiple vendors, versus the model labs which are limited to their very own systems,” Yang said in a recent interview with VentureBeat. “We are also developing our own proprietary technology to seamlessly integrate all the things.”
From Uber to Tesla: How professionals are saving hours every week with browser automation
The first customers are reporting significant increases in productivity through the automation of previously manual processes. Kailiang Fu, a product manager at Uber, said Composite “takes off my shoulders what used to cost me hours of labor per week, corresponding to updating project trackers and running data queries.”
Yang described a security engineer at a big technology company using Composite to automate weekly security architecture reviews – a process that previously required collecting information manually Girub, confluence, Google Driveand internal dashboards before being summarized into reports.
“It’s a extremely, really lengthy process,” Yang explained, “since it uses a variety of different internal tools that don’t have APIs.” The engineer now uses a single prompt to perform composite research across all of those systems and create the vital documentation.
Balancing AI monitoring with enterprise security: Composite's privacy-first approach
The prospect of AI systems monitoring all browser activity raises significant privacy and security concerns, particularly for enterprise customers handling sensitive data. Composite has developed several protections to deal with these issues, including blacklisting for sensitive sites, explicit user confirmation for high-risk actions, and opt-out options for data collection.
“We take data security, privacy and security very seriously,” Yang stressed, noting that Anthropic’s investment reflects shared values ​​around AI direction and security.
The system maintains user control by running in visible browser tabs quite than on distant servers, allowing employees to simply stop or override automated actions. “It's extremely straightforward so that you can close it, stop it and interact with it at any time since it's in your browser,” Yang said.
Viral Growth and Enterprise Adoption: Why Composite Gained Thousands of Users in Two Months
Composite's rapid growth – reaching hundreds of users across lots of of corporations in only two months – suggests strong market demand for browser automation tools. The company primarily experiences organic growth through word of mouth throughout the organizations.
“We have a user at a really large tech company who got their entire team using it, and it's being distributed to different teams across the corporate,” Yang said. This viral adoption pattern sparked company inquiries from managers who saw multiple team members using the tool.
The timing seems good as corporations increasingly search for AI solutions that integrate with existing workflows quite than requiring wholesale technology changes. According to Gartner, greater than half of employees say they spend an excessive amount of money Spending 50% of their workday on repetitive tasks – Copying data between systems, managing email, and updating project trackers, while 25% say these tasks take up greater than 75% of their workday.
Why Composite believes it will possibly beat Google and Microsoft at their very own game
While tech giants like Google, MicrosoftAnd OpenAI Composite's investors imagine the startup's focused approach offers advantages as they develop their very own browser AI capabilities.
“There are large incumbents who don't need to do what's best for his or her agents because they should protect their ads and wish eyeballs to manually view their ads,” Yang argued. “None of them are really focused on what we're attempting to do, which is to focus maniacally on being one of the best predictor and executor.”
Menlo Ventures' Kraning sees the competition otherwise: “I don't need to overshadow Yang and Charlie an excessive amount of, but we won't be selling a tool, we'll be selling results and a system that changes the way in which people and organizations work.”
Technical hurdles and scaling challenges in long-running AI automation tasks
Despite its early success, composite faces technical challenges related to lengthy, complex tasks. Yang acknowledged that automating workflows that require hundreds of sequential steps – corresponding to processing lots of of support tickets – stays difficult resulting from the contextual limitations of the AI ​​model and the necessity for parallel processing.
“How can we parallelize this? How can we run these 400 processes at the identical time after which bring them back together? How can we maintain scheduling for all of them?” Yang said, describing current technical priorities.
The $5.6 million funding will speed up product development and go-to-market efforts, with plans to each enhance single-user capabilities and add enterprise collaboration features.
The way forward for work: From routine digital work to human potential
Yang's ultimate vision goes beyond productivity tools and goals to fundamentally change the way in which people experience work. “Most people spend their working hours at their desk doing fairly strenuous work. They watch the clock until 5 p.m. so that they can go home and do things they really enjoy,” he said.
“We need to help people get into the flow state and really reach their human potential. I feel that's a extremely powerful mission.”
The company plans to expand Windows support and introduce recent personalization features that proactively discover and automate repetitive tasks without explicit user requests. Yang envisions a system that knows users so well that it will possibly predict their needs: “It's like an ideal friend – you will have to know exactly what you would like before you even tell us.”
As corporate interest grows and tech giants mobilize their resources, Composed faces the classic startup challenge: scaling quickly enough to determine a market position before larger competitors can repeat their approach. The company's decision to increase existing browsers quite than replace them offers a realistic path to adoption, but additionally leaves them vulnerable to platform changes Google, MicrosoftAnd Apple.
The stakes transcend the business success of composite. If the corporate's vision proves correct, the character of data work could change dramatically – from people drowning in digital hassle to AI doing the mundane, while people deal with creativity, strategy and real problem solving.
For hundreds of thousands of execs who check their email at 4:55 p.m. and dream of more meaningful work, that future can't come soon enough.
