Frekil Compresses Real-World Evidence from Months to Minutes for Life Sciences Teams
Frekil, a Y Combinator-backed startup, is on a mission to collapse the timeline for generating real-world evidence in the pharmaceutical and biotech industries — from months to minutes. In sectors where regulators rely on real-world evidence (RWE) to confirm drug safety, payers use it to negotiate pricing, and manufacturers need it to expand indications and prove outcomes, the stakes could not be higher. Yet generating RWE today still typically takes months — long enough that safety signals go unexamined, approvals stall, and life-saving insights arrive well after they could have made a difference.
Based in San Francisco, Frekil accelerates real-world evidence generation for life sciences companies by turning the traditional multi-stage RWE workflow into an AI-driven pipeline. The platform connects to fragmented clinical data — electronic health records, medical claims data, and other real-world sources — harmonizes that data into analysis-ready form, and lets teams run end-to-end studies from cohort design to statistical analysis plan (SAP), hypothesis testing, and final reports. The company is adamant that speed cannot come at the expense of scientific rigor, and emphasizes full auditability throughout the workflow so that regulatory-grade evidence remains defensible at every step.
The founders, Nikhil Tiwari and Shivesh Gupta, both graduates of IIT Bombay, have assembled an unusual combination of healthcare research background and high-stakes software engineering experience. Tiwari, the co-founder and CEO, previously worked as a software engineer at Stripe, Amazon, Marsh, The Arena, and Truscroll, and holds an exchange semester credential from the University of Geneva that included work at CERN. He also pursued healthcare AI research at IIT Bombay. Gupta, the co-founder and CTO, worked as a systems software engineer at Sony Japan and declined a full-time offer from a high-frequency trading firm to build Frekil, drawing on his leadership of the institute’s web and coding club during his time at IIT Bombay. The pair bring the systems discipline required to handle messy, sensitive clinical data at scale alongside the domain seriousness the life sciences industry demands.
The market is significant. Life sciences companies collectively spend billions of dollars a year on RWE studies, and the strategic importance of those studies is only growing as regulators increase their emphasis on post-market surveillance and outcomes-based pricing gains momentum. Every major pharma company has dedicated RWE teams, and every one of them is grappling with fragmented data sources, long analyst turnaround times, and study-to-study inconsistency. Frekil’s bet is that a purpose-built AI platform — one that understands clinical data structures, preserves full auditability, and automates the repetitive layers of study execution — can deliver order-of-magnitude improvements in speed without compromising defensibility.