Backend Engineer - Applied Agents

A unified OS for drug development: our investment in Cheiron. July 22, 2026 Drug development is one of the hardest problems in the world. It takes a decade, costs billions, and fails 90% of the time. Much of this is a natural consequence of how complex these programs are. A drug in development touches clinical, regulatory, manufacturing, and commercial functions all at once, generating knowledge across each that rarely makes its way into a single shared picture. Even the most talented scientists and program leads end up spending a significant amount of their time hunting down context rather than acting on it. The result is slower timelines, decisions made on incomplete information, and drugs that take longer to reach the patients who need them. This is the problem the Cheiron team set out to solve. Today, we're thrilled to announce Cheiron's $8 million seed round, led by Menlo Ventures, with the backing and strategic support of industry veterans including Moderna co-founder and MIT Institute Professor Robert Langer; former Pfizer Chief Medical Officer Freda Lewis-Hall; Chai Discovery co-founder and CEO Josh Meier; former Starbucks CEO Laxman Narasimhan; and former Apple AI chief John Giannandrea. A drug program, represented as a whole. If we look at what it takes to advance a drug program, at any given moment, a team is simultaneously tracking efficacy signals, monitoring the safety profile, stress-testing the regulatory strategy against precedent, and watching competitors for anything that might invalidate their assumptions. At the heart of this is a single argument - that this molecule treats this disease, safely and effectively - which has to hold up at every stage and is being made across dozens of people, hundreds of documents, and years of accumulated context. Yet none of it is connected. Cheiron built the infrastructure to piece it all together. The platform's Lifesciences Knowledge Graph (LKG) ingests and structures biomedical, clinical, regulatory, patent, and commercial knowledge, then maps it against the specific contours of a customer's program. The result is a drug program you can actually query. For the first time, teams have a single place to see the full state of their program and stress-test it in real time. Is the safety narrative still consistent with the latest data? Does the trial design hold up against recent regulatory precedent? Are there gaps in the evidence package that need to be closed before a submission? Cheiron holds the state of the program so experts can focus on judgment, and every answer is traceable to a source and defensible in a room full of scientists. Cheiron's early traction speaks for itself. In less than six months since launch, the platform has reached more than 20% of Korea's biopharma knowledge-worker base and secured seven of the country's top 10 biopharma companies as customers. Building on this momentum, Cheiron is now running active pilots with leading global pharmaceutical companies as it expands into the U.S. market. A customer-obsessed team. The best founders understand their customers at a level that borders on obsessiveness, and Minseok Bae, Jason Park, and Harshit Gupta have that quality in abundance. They met at Stanford and bring extraordinary technical pedigree across competitive programming, biology, and physics, but what stayed with us after our first meeting was how clearly they saw the problem from the customer's seat, and how relentlessly they'd already moved to solve it. The team has also assembled a deep advisor bench of industry veterans across pharma who have led FDA approvals and built global drug programs from the ground up. Backing the Future of AI and Biology. At Menlo, we've long believed that one of AI's most transformative applications lies at the intersection of AI and biology. Cheiron joins a portfolio we've been deliberately building at this intersection, including Chai Discovery, which is developing foundation models for biology, Phylo, building an IDE for biological design, and Aurora Therapeutics, developing therapies for rare diseases. We're thrilled to welcome Minseok, Jason, Harshit, and the Cheiron team to the Menlo family.

Cheiron, a Los Altos-based startup building an operating system for drug development programs, has raised $8 million in seed funding led by Menlo Ventures. The round included backing from notable industry figures including Moderna co-founder Robert Langer, former Pfizer Chief Medical Officer Freda Lewis-Hall, and former Apple AI chief John Giannandrea. The company's platform uses AI to model drug programs as connected systems, consolidating scattered information from experiments, clinical results, regulatory interactions, and scientific literature. At its core is the Life Sciences Knowledge Graph, which structures biomedical, clinical, regulatory, patent, and commercial knowledge into a single model. Since launching less than six months ago, Cheiron has been adopted by tens of thousands of biopharma professionals. The funding will support expansion of the platform's capabilities across clinical, regulatory, and strategic workflows, and grow the company's engineering and life sciences teams.