A bit about Chetan's background. He joined EvolutionaryScale as their first product engineer, launching an AI inference platform to help scientists design proteins. He joined Instabase as #16 helping banks with document & imaging processing, closing $7m as the technical lead. When looking to start his own company, Chetan realized there are many AI startups focused on drug discovery, but a lot less attention bringing those products to market.
😖 The Problem: Knowing What The FDA Thinks
The FDA is the customer of every function in life sciences. It doesn’t matter whether you are regulatory affairs, R&D, quality, manufacturing, commercial, etc.
At the same time, it’s difficult to understand their thinking. Guidance documents are high level and vague. The only alternative is doing (or paying a consultant to do) manual, painful research on past decisions they’ve made and understand why. Other regulators around the globe are no easier.
ChatGPT / Claude / etc. have only made it worse:
They trust their knowledge and only look up what they feel the need to. This means low-latency, but high-hallucination research.
Even when they retrieve sources, at most, it’s a few dozen. It’s also liable to include lots of internet junk mixed in with the authoritative government databases the industry trusts.
They don’t make it easy to check citations. They cite a 1200 page PDF and you don’t really know where to dig inside it.
✨ The Solution: Hallucination-Free, Deep Research For Life Sciences
Rhizome solves this problem by providing hallucination-free and thorough deep research for the life sciences. There’s a few key things their customers like:
Rhizome didn’t have a single hallucination reported in 2025.
Every statement comes with a citation.
Rhizome reads up to 1,000 documents per question in a few minutes before writing an answer. They’re continuously extending this limit.
Lorem ipsum dolor sit amet, consectetur adipiscing elit. Suspendisse varius enim in eros elementum tristique. Duis cursus, mi quis viverra ornare, eros dolor interdum nulla, ut commodo diam libero vitae erat. Aenean faucibus nibh et justo cursus id rutrum lorem imperdiet. Nunc ut sem vitae risus tristique posuere.