Magentic raises $18M to deploy AI workforces in global manufacturing
Magentic, a London and New York-based AI startup founded by alumni of McKinsey and OpenAI, has closed an $18 million Series A to scale autonomous digital workers across the procurement and supply chain operations of large industrial companies. The round was led by Felicis, with continued backing from Sequoia Capital and The Westly Group, arriving just twelve months after the company's July 2025 launch.
The funding positions Magentic at an unusually sharp intersection: the physical economy's mounting capital pressures, on one side, and the rapid maturation of multi-agent AI systems capable of operating inside legacy enterprise infrastructure, on the other.
The physical economy's AI inflection point
The timing is deliberate. Goldman Sachs, the company says, projects roughly $8 trillion in AI-related capital expenditure between 2026 and 2031, a significant portion of which flows into physical infrastructure that must be sourced, contracted, and built. At the same time, procurement workloads at large manufacturers have grown approximately 10% year on year, against budget growth of just 1%. That mismatch is the market Magentic is targeting.
Rather than layering a software dashboard over existing systems, Magentic's digital workers operate natively inside a manufacturer's own environment, communicating through Microsoft Teams and email, and interfacing with ERP systems and fragmented data estates that have accumulated across decades. A single customer is said to now process more than one million orders annually through Magentic's agents; another has recorded $4 million in verified savings. Across a reported base of Global 500 companies, including three of the world's ten largest beverage producers, the company claims typical outcomes of 2–5% procurement savings, a 60% improvement in data quality, and a reduction of tens of thousands of manual working hours.
"Supply chains are the least glamorous part of the economy, yet the most consequential, deciding what gets built and what does not," said Feyza Haskaraman, Partner at Felicis. "Getting an agent to understand a manufacturer's complex systems well enough to take action inside them is no small feat."
Cross-sector implications for capital and automation
For cross-sector investors, Magentic's raise is a signal worth reading beyond its headline figure. The industrial AI agent space has so far attracted most venture attention at the hardware and robotics layer, autonomous mobile robots, computer vision for quality control, AI-assisted equipment maintenance. Magentic represents a different bet: that the highest-value automation opportunity in the physical economy is not on the factory floor but inside the procurement office, where decisions about raw materials, supplier selection, and contract negotiation compound across billions in annual spend.
This matters for the broader capital landscape. As the AI infrastructure buildout intensifies demand for physical components, semiconductors, steel, rare-earth materials, data-centre cooling equipment, the procurement function sitting between that demand and global supply chains becomes structurally more critical. Manufacturers exposed to tariff volatility and geopolitical supply disruption have limited margin for slow, manual sourcing cycles. Autonomous procurement agents that can operate at scale across fragmented, decades-old systems represent a form of operational resilience that the current trade environment makes urgent.
The security architecture Magentic has built reflects enterprise caution about agentic AI in sensitive systems. Zero-data-retention agreements with AI model providers, isolated cloud deployments, and region-specific data residency are now table stakes for any vendor seeking to operate inside Global 500 infrastructure. That compliance posture also has implications for the regulatory trajectory of agentic AI more broadly: as vendors demonstrate that autonomous agents can function within enterprise governance frameworks, the policy conversation around AI liability and auditability in critical industrial operations is likely to accelerate.
New capital will fund an expanded research agenda focused on long-horizon AI reasoning, enabling agents to work across terabytes of multimodal data simultaneously, a prerequisite, the company argues, for tackling the most complex optimisation problems in direct spend categories such as raw materials. The direction of travel points toward a future where procurement teams are smaller in headcount but larger in strategic reach, a shift that will ripple into workforce planning, ERP vendor roadmaps, and the economics of the management consultancies that currently fill the gap.