Inveta raises $311M to map life's chemical code and turn it into blockbuster drugs

Inveta raises $311M to map biological compounds into drugs

Inveta has raised $311 million to advance a platform it describes as a sequencer for life's chemical code. The company's core argument is that of the estimated 1 to 10 billion biological compounds that exist in nature, only around 400,000 have ever been identified — leaving 99% of the chemical makeup of living organisms unknown to science. Inveta is building technology to take a biological sample and answer two questions: what are the molecules, and what do they do?

Viswa Colluru, Inveta's founder and CEO, is explicit about the company's positioning. Rather than selling software to pharma or building general-purpose AI models, Inveta is developing and owning its own drugs. His rationale is blunt: every large company in the industry has been built on blockbuster drug franchises. He points to Eli Lilly, where roughly 70% of enterprise value derives from forward sales of the GLP-1 family, Novo Nordisk at a similar 80%, and Sanofi at 30% from a single drug — Dupixent, for eczema. The strategic logic follows from those numbers: make drugs that matter, or don't bother.

“About 400,000 compounds have been discovered by the collective human endeavor throughout history to now. And it's expected that there's about one to 10 billion. So 99% of what makes up you, a tomato in your garden, or a random sample in the Amazon Rainforest is still a mystery to science... The first one is: make drugs and make drugs that matter.”

First two programs

Inveta's initial pipeline targets metabolic disease and inflammatory conditions rather than cancer. Its first molecule aims to treat atopic dermatitis and asthma with a safe, non-steroidal oral — a category that hasn't seen a new approved option in over 25 years. The second program centers on a hormone the company says the body produces after intense exercise. Inveta has formulated it into a once-daily pill, positioning it as a maintenance option for people who come off GLP-1 drugs. Colluru estimates 55 million Americans will have taken a GLP-1 over the next seven years and then face no clear next step — a gap Inveta is explicitly targeting.

He notes that 10 to 15% of patients don't respond to GLP-1s at all, likely driven by hormonal differences, and that the biology of common diseases like obesity involves multiple organs — fat tissue, brain, muscle, pancreas — coordinating in ways that vary significantly across individuals. That complexity is what makes these conditions hard for AI, not just for conventional drug discovery.

AI and biosafety

On the question of whether AI-designed molecules pose safety risks, Colluru offers a more grounded assessment than the public debate usually allows. Inveta's models are trained to characterize what evolution has already produced, not to design novel compounds from scratch — which sidesteps many of the more speculative biosecurity concerns. For models that do generate new molecules, he argues the real constraint is synthetic feasibility: more than 99% of AI-designed chemistry is physically impossible to build given the energetics of synthesis, which limits how quickly anything dangerous could be produced. The harder bottleneck, in his view, is that AI remains poor at manipulating the physical world, and complex biological workflows still require custom, hard-to-automate processes. The more plausible risk, he says, involves a human actor deliberately in the loop with an AI system — a social and governance problem more than a technical one.

The $311 million gives Inveta the runway to push both programs through clinical development. Whether the platform's underlying discovery engine — mapping unknown biological compounds at scale — translates into a repeatable drug pipeline is the question the capital is meant to answer.