Startup Persist AI, which launched a cloud lab to accelerate pharmaceutical formulation development, raised $12 million in a Series A funding. Announced on Wednesday, the funding round was led by Spero Ventures with participation from MBX Capital, Shimadzu Future Innovation Fund managed by Global Brain Corporation, Eli Lilly & Company, SignalFire, Ford Street Capital, Purdue Ventures, Good AI Capital, Cartography Capital, Asymmetry Ventures, and existing investors 2048 Ventures, Innospark, and Y Combinator.
“Every drug that reaches the market depends on an optimal formulation. While the industry has heavily invested in AI and predictive tools across the drug development pipeline, formulation has remained a blind spot—until now,” said Sara Eshelman, General Partner at Spero Ventures.
READ: Document extraction startup Reducto raises $24.5 million in funding (April 29, 2025)
“Persist dramatically reduces both the time and cost at every stage of development, ultimately lowering the hurdle for investment in next-generation therapeutics like long-acting injectables.”
Persist AI makes use of an AI-driven robotics platform that helps reduce drug formulation development time from years to months. The startup has identified several issues linked to formulation including process complexity, scale-up challenges, material constraints, and data limitations.
Founded by Karthik Raman and Christopher Shelner, Persist AI addresses these challenges using a platform built on three foundational pillars — custom automation for chemistry rather than just biology, miniaturization with scale-up built in, and AI that learns from data.
The company claims it has custom robotics specifically designed to handle organic solvents and complex materials used in pharmaceutical formulations, allowing them to generate formulation data 30 times faster than traditional methods. It also makes use of microfluidics technology to miniaturize formulation development, requiring minimal drug material while creating precisely controlled particles.
READ: AI chip startup FuriosaAI turns down $800 million acquisition offer from Meta (March 24, 2025)
This approach is unique because Persist AI’s microfluidic system is designed for seamless transition from small-scale discovery to large-scale manufacturing, eliminating the traditional scale-up challenges that delay drug development.
Lastly, the data generated by the company’s robotics is comprehensive, structured, and linked to manufacturing process parameters, allowing it to build AI models that can predict crucial properties like drug loading and release characteristics with remarkable accuracy.
Persist AI plans to make use of its Series A funding to deploy a cGMP-compatible manufacturing line to fulfill customer contracts, while continuing to build its cloud lab platform.