Artificial intelligence agents are becoming increasingly capable of carrying out digital tasks, but a major obstacle remains: much of the internet was built for people, not autonomous software. From downloading invoices to researching competitors, many everyday business activities still require agents to navigate websites manually, clicking buttons, filling forms and moving between pages. French American startup Reduck is developing a faster alternative, raising $3 million in pre seed funding to help AI agents interact with websites and business applications without relying on traditional APIs.
$3 Million Funding Round
The funding round was led by Daphni, with participation from Drysdale Ventures and Galion.exe. The investment will support Reduck’s efforts to improve its browser automation technology, expand its library of ready to use scripts and introduce new tools for developers building AI powered applications.
Founded by CEO Daniel Huynh, CTO Louis Abraham and COO Raphaël Millet, Reduck is targeting a growing challenge in enterprise automation: enabling AI agents to work with digital systems that lack direct software integrations.
Why AI Agents Need Better Website Access
Many business applications provide application programming interfaces, commonly known as APIs, that allow software systems to exchange information directly. However, numerous websites and specialised business platforms lack suitable APIs or Model Context Protocol interfaces that AI agents can use.
Without these connections, agents often have to interact with websites through a browser in the same way a human would. They may need to locate buttons, enter information, navigate menus and repeat the same steps across multiple pages.
This approach can become slow, expensive and unreliable, particularly when agents perform complex or repetitive tasks.
Reduck wants to reduce this friction by giving agents access to reusable browser automation scripts that can perform specific tasks more efficiently.
Reduck MCP Turns Browser Tasks Into Reusable Workflows
The company’s platform, Reduck MCP, allows AI agents to discover, execute and create browser automation scripts. Instead of navigating a website from scratch for every request, an agent can use an existing script designed for the required task.
According to Reduck, its technology can make agents three times faster and cheaper across supported tasks. Examples include collecting invoices from multiple vendors and conducting market research on LinkedIn.
Its existing library contains more than 1,200 ready to run automation scripts covering over 250 websites. This collection gives developers a starting point for automating tasks across different online services without having to build every workflow themselves.
The platform is designed to make website interactions more repeatable while reducing the computational resources needed to complete common operations.
AI Agents That Build Their Own Automations
Reduck is also working on technology that allows AI agents to create and maintain automation scripts autonomously.
This capability could help the platform expand beyond a fixed library of predefined workflows. As websites change their layouts or processes, automation scripts may need to be updated to remain functional. Reduck is focusing on reliability and scalability as it develops systems to manage this challenge.
The company is also preparing a developer offering for teams that want to connect AI agents to applications without APIs. Potential use cases include specialised insurance and healthcare portals, internal customer relationship management systems and other business tools.
Expanding Into Enterprise Automation
Beyond browser automation, Reduck is exploring opportunities in knowledge work, including generative engine optimisation, an emerging discipline focused on improving how brands and content appear in AI generated answers.
The new funding has supported team expansion across AI engineering, full stack development and infrastructure. Further investment will help the company refine its technology, broaden website coverage and launch its developer focused offering.
What Comes Next?
As businesses move from AI experimentation towards automated workflows, connecting agents to existing software remains a practical challenge. Reduck is betting that reusable browser automation can bridge the gap between intelligent AI models and the websites where much of business activity takes place.
Its next challenge will be maintaining reliable automation across changing websites while meeting the security and performance requirements of enterprise customers. If successful, Reduck could help make autonomous agents useful across a wider range of digital tasks, including those that traditional software integrations cannot easily support.