INXM Raises $6.2M for AI Process Engine

Jan de Vries · 2026-06-03

Berlin-based startup INXM has emerged from stealth mode with a $6.2 million pre-Seed round to build what it calls the first AI process execution engine for enterprises. The company aims to solve a problem many businesses know all too well: AI projects that take years, cost millions, and still break more than they fix. The round was led by Cherry Ventures and Redstone, with support from Angel Invest and other backers like Linden Capital. INXM plans to use the funding to launch its first enterprise deployments. ### The Problem with Enterprise AI You've probably seen it happen. A company spends months or years on an AI implementation, hires a team of engineers, and ends up with a system that creates more work than it saves. Knowledge workers still spend their days copy-pasting data between ERP systems, PLM tools, Excel spreadsheets, email, and approval workflows just to close out a month. "We founded INXM because we've seen first-hand how enterprise AI projects fail," says Alex Oelling, CEO of INXM. "Years of implementation, armies of engineers, and AI systems that break more than they fix." The goal is simple but ambitious: build AI that actually finishes the work for you. ### What Is Compiled AI? INXM is built on a new architectural principle they call Compiled AI. Instead of asking a large language model (LLM) to interpret every single transaction in real time, the system uses AI to design and improve processes first. Then it runs those processes deterministically. Here's how it works: - The INXM Orchestrator takes user intent and turns it into executable Plans - Those Plans coordinate work across systems, people, and processes - The engine executes them the same way every time, producing repeatable, auditable outcomes "At its core, Compiled AI means you use LLMs to generate deterministic, enterprise-ready code," explains Matthias Kainer, CTO of INXM. "You then run the code to achieve your outcome. This gives you the flexibility of natural language from AI models, but the testability of deterministic code." ### Built for Real-World Operations What sets INXM apart is its focus on reliability. Most AI tools are great at generating ideas but terrible at executing them consistently in a business environment. INXM's Orchestrator gives operations teams the predictability they need and compliance teams the audit trail they require. The company emphasizes that this isn't another costly tool meant to rip and replace existing systems. Instead, it coordinates your current technology stack to enable collaboration. The result? Processes that can be reliably automated within a few months. "Founders who have brought rocket engines and air taxis to production readiness develop a different mindset than most software teams," says Michael Brehm, Founding Partner at Redstone. "They've seen firsthand what it takes to achieve greatness under real conditions." ### European Roots, Global Ambitions INXM is built in Europe with full data ownership and local deployment. That means it's designed from the ground up to meet the compliance and governance requirements of European enterprise customers. Founded in 2025 by Oelling, Matthias Kainer, Jesper Bylund, and Kamil Kluber, the team brings deep experience from industries where failure isn't an option. Their background includes everything from rocket engines to air taxis, which gives them a unique perspective on building systems that work under pressure. The company believes that AI workflow challenges almost always trace back to brittle integration with day-to-day operations. INXM is built by people who've lived that reality. ### What's Next? With $6.2 million in the bank, INXM is ready to power its first enterprise deployments. The company plans to prove that AI can move beyond being just a productivity tool and become the operational backbone of European industry. If they succeed, the days of copy-pasting between systems might finally be numbered.