An estimated 77% of the world’s transactions touch an SAP system at some point. And most of that code is decades old, written in a proprietary language, and locked inside the workflows at the largest enterprises on earth. It is also, by SAP’s own mandate, on a clock.
Nova Intelligence is betting that’s a generational opening. The San Francisco-based startup has raised a $31.5 million Series A led by Chemistry, Fortune exclusively learned, bringing total funding to more than $40 million. Existing investors Accel, Conviction, and SAP.io—SAP’s venture arm—also participated.
Nova builds an agentic AI platform for SAP—analyzing, modernizing, and generating the custom code that runs payroll, supply chains, and finance functions inside Fortune 500 environments.
“We’re an AI platform for SAP that accelerates the work that companies are doing on SAP across really the full life cycle, from design to development to testing,” co-founder and CEO Emma Qian told Fortune.
Timing is a key part of Nova’s thesis. SAP has set a mandatory migration deadline for customers to move from legacy systems to its modern S/4HANA platform—recently pushed from 2027 to 2030—and the surrounding services market is enormous. Analysts peg the broader S/4HANA opportunity at more than $89 billion when implementation, upgrades, and ongoing support are included. “We’ve met companies that are budgeting over a billion dollars just for this migration,” Qian said.
Qian, a former research engineer at Google DeepMind and Meta AI, founded Nova alongside repeat operator Sam Yang and Professor Alexander Zeier—the co-inventor of SAP HANA and former CTO of Accenture’s SAP Business Group. Justin Kershaw, formerly CIO of Cargill, a Fortune 25 company and one of SAP’s largest customers, joined as chief customer officer. The company is still just 20 people.
For Chemistry cofounder Kristina Shen, who led the round, the 2030 migration deadline is only a wedge. “The daily maintenance, development, and operations use cases mean Nova becomes the platform SAP teams rely on indefinitely, and eventually the system of action for all critical systems,” she told Fortune.
Nova’s bet is that agentic AI can compress that work dramatically. An April 2025 case study with Kyndryl showed Nova’s system led to a 75% reduction in manual effort and a 50% cost reduction of modernizing a company’s SAP platform. New customer numbers are tracking similarly. Festo—one of the first 30 SAP customers in the world, running the system since the 1970s—estimated they can work about five times faster, Qian said. “Some really complex programs used to take months to do, but now they take a single day,” she added.
The competitive question is what happens when SAP itself builds more of this technology in-house. Qian’s answer leans on alignment. “Our goal is to get customers to a state of their SAP system that’s as good and as optimized as possible,” she said. “That’s where we think we’re actually pretty complementary.”
The new capital will fund both R&D and a meaningful expansion of Nova’s go-to-market team targeting flagship enterprise logos. Long term, Qian sees Nova extending beyond SAP. “Your business process doesn’t typically live within one system,” she said. “Our goal is to expand to the other commonly used systems that customers are using as well.”
For Chemistry, it’s a bet on a vertical agent in a market the generalized AI giants can’t easily touch. For Qian, the pitch is simpler: the world’s most critical software is finally getting an AI-native co-pilot.
