Our objective is to deliver first-in-class medicines against disease targets that cannot currently be drugged effectively or at all. Therefore, we are building a lights-out wonder drug factory, where AI runs the entire design-make-test loop. Linking superhuman judgment to robotic throughput is the path to reindustrializing our life science ecosystem, not through throwing human labor at the problem.

Our team has been working on this for over a decade, so we know that there has never been a better time to build this system.  Abe co-founded Atomwise, one of the first AI-for-drug-discovery companies.  We raised $220 million; hired 100 people; signed over $5B in total deal value with big pharma companies like Sanofi, Bayer, Eli Lilly, and others; and ran the biggest application of AI to hit discovery in history. Saulo solved the synthesizability problem in generative chemistry AI and wrote structural biology software that remains the standard toolkit (KVFinder; qFit3, STCRDab). Cam co-founded Xbox & Xbox Live (Microsoft) and has been a CEO, CTO, COO, and CPO across AI, biotech, healthcare, and consumer tech. 

Today, we see the convergence of critical enabling computational, physical, and economic trend lines. Agentic systems have become proficient orchestrators and tool integrators, as the frontier labs continue to extend the time horizons over which agents can operate; in our own hands, we have already compressed compound selection timelines 10x by replacing manual processes by agentic-run. The algorithmically-accessible chemical space has been growing exponentially for two decades, and the rise of cloud labs enables closing and scaling the design loop. Finally, the looming Big Pharma patent cliff means that the ecosystem is aligned on the value of innovative new assets. 

At WonderTx, we’re beginning by replacing needles with pills. Many drugs, like insulin or Ozempic, are great except you have to inject them. We discover new molecules that you can take as a pill instead. Because the pills operate on biology that has been shown to be effective in people, rather than merely in animal models of disease, we avoid what has historically been the single biggest risk in drug discovery.  This should dramatically increase our probability of success.

There’s $200B of treatable diseases and validated human biology without a pill because discovery programs lack a tractable chemical starting point, which makes this a zero-shot inference problem.  Because these targets have no training data, off-the-shelf LLMs and other historical interpolative AI approaches are insufficient. Instead, we are building our own extrapolative frontier models that reason beyond their training data. Most critically, we have repeatedly validated our platform on both our own and partner programs. On 4 biologically- and structurally-diverse targets where we had zero training data, we successfully picked binders that were validated experimentally, with confirmation ranging from primary screen hits to crystallographic structures.