Nearly three years after the ChatGPT moment, companies are moving from next-token prediction (text-in-text-out), to automating the work and delivering outcomes (text-in-action-out). With that comes another change in the world of enterprise sales: companies move from selling ROI through seats to selling ROI through outcomes. At Sequoia, we are very excited about the opportunity for vertical agents to deliver outcomes—not only by going after software budgets, but also by taking a cut of the savings that LLMs and thoughtful, domain-specific tailoring unlock. 

If sales is responsible for customers, procurement is responsible for suppliers—and for most companies, the vast majority of costs flow through that function. Procurement teams work tirelessly to ensure suppliers deliver on time, at the right price and in the right quantities. But here’s the challenge: large enterprises work with *tens of thousands* of suppliers while their procurement function is 1/100th of that size. This is a costly gap. McKinsey found that for large enterprises, unfulfilled supplier obligations can leak roughly 2% of spend—for many customers, that’s tens of millions wasted every year.

In an ideal world, each supplier would have a dedicated employee reviewing every contract clause, inspecting every shipment, verifying every invoice and recovering leakage when deliveries fall short. That world has never existed—until now. 

Magentic has created AI agents they call “Mages,” which integrate into existing ERPs, partner seamlessly with procurement teams and immediately impact P&L. Mages are just as magical as they sound: they autonomously search thousands of documents and millions of deliveries to identify late payments, manufacturing quality concerns and other potential issues, then summarize those problems—and propose solutions. Mages can find documents, clauses and rules and track the entire chain of events that led to an issue. Then, they will suggest actions for human review, like drafting emails, updating invoices or notifying stakeholders.

Procurement team members exponentially increase the scale of their work, all while keeping full control. Whether the goal is recouping costs or simply streamlining their supply chain to ship products faster, Magentic can help. This product is not taking existing work done by humans and replacing it with AI. It’s doing work that cannot be done by humans, to achieve both savings and supply chain resilience.

Co-founders Robin Van Aeken and Odhran O’Donoghue met while winning hackathons at Oxford and arrived at this idea from complementary backgrounds. Odhran comes from the bleeding edge of AI research at OpenAI, and Robin understands the world of procurement having worked at McKinsey & Company. Robin and Odhran are forces of nature who joined us last year for Arc—Sequoia’s company-building immersion for pre-seed and seed-stage founders—and we were privileged to have them start the company from our office. We are proud to double down on our support for Robin, Odhran and their growing team by leading their seed round.