Arrakis surfaces from stealth mode with $38 million to integrate AI into factories and supply chains.

      TL;DR: Arrakis has secured $38 million, with Blossom Capital and Accel leading the funding, to develop an AI operating system aimed at industrial sectors like aerospace and logistics.

      Arrakis, a startup based in London and Paris that focuses on creating an AI operating system for industrial enterprises, has revealed its operations after raising a total of $38 million in funding. The company believes that significant returns from artificial intelligence will originate from factories and supply chains rather than office software. This funding includes a $30 million Series A round led by Blossom Capital, in addition to a $7.5 million seed round led by Accel that concluded in March. Fortune, which initially reported on the funding, stated that the Series A values Arrakis at $140 million post-money.

      The company was co-founded in January 2026 by CEO Rafael Quintanilla—who was previously a vice president at Accel—alongside Haroun Beltaifa and Romain Fouilland, both ex-Palantir employees, and Mikhail Galkov, former employee of Delivery Hero. Quintanilla has noted that the founding team’s connection to Palantir is intentional, viewing it as a legacy entity that is prime for disruption. Arrakis presents itself as a more agile and competitive alternative to Palantir's established government and enterprise contracts, as well as to traditional consulting firms that charge based on headcount rather than outcomes.

      The startup utilizes what it terms forward-deployed AI engineers who integrate directly at customer sites in industries such as aerospace, energy, logistics, and manufacturing. Instead of locking clients into a specific model provider, Arrakis adopts a model-agnostic strategy, initially using commercial models from OpenAI and Anthropic, then transitioning workloads to open-source options from companies like Mistral, which has recently enhanced its capabilities by acquiring the industrial AI firm Emmi. Arrakis asserts that this approach can enhance output quality by two to four times while reducing token costs by around 70 percent.

      Arrakis has already secured five paying clients, including publicly listed companies, and one of its clients reportedly reduced procurement cycle times by 90% using the platform. The company links about half of its fees to performance outcomes, a distinctive model within enterprise software that Quintanilla claims demonstrates the system's ability to provide tangible results. The startup has found favorable traction among family-run businesses, which generally make quicker purchasing decisions compared to publicly traded firms with complex approval systems.

      The roster of angel investors adds credibility in both AI and industrial sectors, including Datadog CEO Olivier Pomel, OpenAI's head of business products Olivier Godement, and Junaid Hussein, founder of Cambridge Aerospace, who all participated in the funding round. Additional support came from GFC, MainObject, and Rerail.

      Arrakis is entering a competitive but rapidly expanding market for industrial AI. For instance, PhysicsX raised $300 million at a valuation of nearly $2.5 billion in June for AI-driven engineering simulation, while Prometheus, backed by Jeff Bezos, is pursuing a similar focus in heavy industry. Quintanilla contends that Arrakis offers a distinct advantage in terms of speed, being able to deploy its solutions within weeks as opposed to the months or years typical of legacy vendors.

      The company intends to triple its workforce from around 15 employees and establish offices in New York and the Middle East, utilizing the new capital to grow its footprint in markets prioritizing industrial transformation. The key challenge for Arrakis will be scaling its forward-deployed model without incurring excessive costs, determining whether its approach will be successful beyond its initial group of clients.