Arrakis emerges from stealth with $38 million to build industrial AI operating system

The London- and Paris-based startup says it targets aerospace, energy, logistics and manufacturing, and was valued at $140 million post-money after a $30 million Series A.

Summary

Arrakis, a seven-month-old startup based in London and Paris, has emerged from stealth with $38 million in venture funding to build what it describes as an AI operating system for industrial companies. The company is targeting sectors including aerospace, energy, logistics and manufacturing, arguing that AI adoption has focused too heavily on desk-based work rather than the operations tied to physical goods. The latest financing was a $30 million Series A led by Blossom Capital, with participation from Accel, GFC, MainObject and Rerail. Accel had previously led a $7.5 million seed round, while individual backers include Datadog CEO Olivier Pomel and OpenAI’s head of business products, Olivier Godement. Cofounder and CEO Rafael Quintanilla told Fortune the new round values Arrakis at $140 million post-money. Quintanilla, a former vice president at Accel, said he founded the company after concluding that industrial businesses were underserved by the current wave of AI products. He positioned Arrakis against players such as Palantir, consulting firms including Accenture and Boston Consulting Group, and Prometheus, saying Arrakis aims to act as an AI layer for operational workflows around core industrial processes rather than replace engineering itself. The company says it begins with narrow operational use cases, such as improving cash-flow visibility, and then expands if customers see measurable gains. Arrakis typically ties about half its fees to hitting a performance target, according to Quintanilla. He also said the startup is designed to be model-agnostic (able to switch between AI model providers), initially using proprietary models from OpenAI or Anthropic before shifting customers to open-source alternatives such as Mistral, or Chinese vendors if permitted, to reduce token costs (fees based on AI usage) and improve output quality. Arrakis currently has five customers and plans to triple its headcount from roughly 15, while opening outposts in New York and the Middle East.

Terms & Concepts
  • model-agnostic: Able to work across different AI models
  • token costs: Fees charged for AI model usage
  • open-source alternatives: AI models with publicly available code