Palette has raised €3 million in pre-seed funding to build an operating system designed for AI-native teams. The round was led by Ugly Duckling Ventures, with participation from Emblem and Acadian Ventures.
Angel investors in the financing include Jesper Brøndum of Boozt, Jonas Gyalokay of Airtame, Niklas Stephenson of Phone.inc, Dennis Kayser of Forecast, Peter Holm of Tonsser, Rasmus Schmiegelow of Goodiebox, Sebastian Brandes, Thomas Hellmann Midtgaard, Jonas Gundersen of Astralis, and Mastra’s Shane Thomas, Sam Bhagwat and Abhi Aiyer, among others.
Palette is developing its platform around the idea that companies should maintain control of their organizational context even as the underlying AI models and tools they use continue to change.
The company defines that context as information including what employees know, decisions teams have made and the processes through which they work. Palette’s approach is designed to allow both employees and AI agents to use that shared context while organizations remain free to switch between different AI providers.
The company’s initial product, Palette Desktop, provides a shared workspace in which employees and AI agents can work from the same files and organizational context.
Teams can use AI systems including Claude Code, Codex, Gemini, Mistral or Palette on company-owned files, select the agent appropriate for a particular task and review changes before they are saved.
Palette said it developed the product after encountering the problem internally while using AI agents extensively. Work remained isolated across separate AI sessions, forcing team members to repeatedly provide the same context while making it difficult for one employee’s AI-assisted work to benefit others.
Palette Desktop is currently available for Mac, with a Windows version planned.
The company plans to use the pre-seed funding to expand its team in Copenhagen, bring Palette Desktop to additional companies and continue developing Palette OS.
Palette OS is being designed as a broader organizational context layer featuring a living context library, shared AI skills, integrations with company tools and handoffs between sessions.
The company’s strategy centers on separating an organization’s proprietary context from any individual AI model, allowing companies to maintain a persistent layer of knowledge while switching among different AI systems as the technology evolves.
KEY QUOTES:
“Most companies are trying to become AI-native, and it’s tempting to chase whichever model is newest. We think that’s the wrong strategy.”
“AI models and tools will keep changing. Your company context should stay in a place you control. That includes what your people know, the decisions you’ve made, and the way you work. People and agents can then use it with whichever provider you choose.”
“Own the context. Swap the model.”
Palette
“We were running agents for almost everything, but the work stayed trapped in separate sessions. We kept re-explaining the same context, and one person’s work rarely helped anyone else.”
“So we built Palette Desktop: a shared workspace where your team and agents work from the same files and context.”
Palette
