Fellou offers an agentic AI browser that lets users automate complex, multi‑app web workflows with a single natural‑language prompt. The platform plans tasks, executes deep searches—including logged‑in sites—and can control desktop applications, generating traceable reports, filling forms, or creating multimedia content while providing real‑time visibility and editability of each step. It also builds a personalized memory of browsing history to deliver context‑aware assistance.
Funding
Funding not disclosed
Founders
Product
Problem
Professionals and knowledge workers spend extensive time manually navigating multiple web services, logging in, extracting data, filling forms, and compiling information from disparate sources. This repetitive “online grunt work” reduces productivity and delays decision‑making.
Solution
Fellou offers an agentic AI browser that transforms a natural‑language prompt into a detailed action plan and then executes the plan across web and desktop applications. The system can scrape data, complete forms, interact with logged‑in accounts, and manage local files, all while providing real‑time visibility of each step. Users can review, edit, or intervene in the workflow before and during execution, ensuring accuracy and control. An integrated memory component learns from browsing history and notes, enabling contextual recall and proactive assistance. The platform delivers end‑to‑end automation for tasks such as competitor analysis, travel planning, multi‑source reporting, and routine digital workflows.
Target Audience
Primary users are knowledge workers, analysts, marketers, and product teams who need to automate complex web and desktop tasks, as well as developers seeking to embed custom AI agents into their workflows.
Features
- Natural‑language intent parsing that generates a step‑by‑step action plan with user‑editable previews
- Cross‑app web automation including data scraping, form filling, and interaction with logged‑in services (e.g., Reddit, X, Notion)
- Desktop integration that controls local applications and manages files for fully system‑level workflows
- Real‑time workflow monitoring with the ability to pause, edit, or override any step
- Agentic memory that indexes browser history and personal notes to provide contextual answers and recall past work
- Multi‑modal output generation: detailed reports with text, images, charts, as well as code, presentations, and media assets
- Scheduling engine for recurring or future‑time tasks and unlimited concurrent task execution
- Extensible framework (Eko 2.0) allowing developers to build custom AI agents using a hybrid of natural language and JavaScript