Onepot AI Raises $13M to Accelerate Drug Discovery

Onepot AI, a pioneering startup leveraging artificial intelligence to streamline chemical synthesis for drug discovery, has announced it has raised $13 million in funding. This capital, comprising pre-seed and seed rounds led by Fifty Years, will fuel the company's mission to accelerate the creation of new medicines by tackling one of the industry's most persistent bottlenecks.

The company was founded by Daniil Boiko, a Ph.D. candidate in machine learning in chemistry at Carnegie Mellon, and Andrei Tyrin, who holds a bachelor's in computer science from MIT. Their collaboration stemmed from a shared frustration within the pharmaceutical industry.

"The best ideas in drug discovery were often blocked not by biology, but by synthesis," Boiko told TechCrunch, explaining that synthesis—the creation of new molecules through chemical reactions—is akin to assembling complex structures from smaller components.

This intricate process of building larger molecules from smaller ones is notoriously difficult and time-consuming. Boiko observed that drug hunters frequently abandoned promising research avenues simply because the required chemical compounds were too challenging to synthesize.

"The compounds never even got a chance to be tested," Boiko lamented.

Tyrin, drawing from his experience in computational pipelines for drug discovery, highlighted the stark contrast between rapid computational modeling and slow lab work.

"The models could generate ideas in hours, but it could take months for the lab to catch up," he noted.

Both founders recognized a critical imbalance: significant investment was flowing into molecular design, while the equally challenging problem of physically creating these molecules was largely overlooked. Boiko also pointed to a geopolitical dimension, citing vulnerable global supply chains and increasing trade competition with China, underscoring the need to rebuild small-molecule synthesis capabilities within the United States.

Onepot AI's Innovative Approach

To address these challenges, Boiko and Tyrin established Onepot AI, centered around its small-molecule synthesis lab, POT-1. They also developed "Phil," an AI organic chemist designed to conduct experimental analysis, thereby accelerating the compound synthesis process for their early commercial partners, which include various biotech and pharmaceutical companies.

Traditionally, pharmaceutical and biotech firms either maintain extensive in-house chemistry teams or outsource to contract research organizations, often overseas. Human chemists can spend months and thousands of dollars to synthesize a single compound, a process fraught with trial and error involving studying compounds, collecting data on biological activity, drug pharmacokinetics, and toxicology.

"The main limiting factor here is not testing these compounds, but making them in the first place," Tyrin explained. "We aim to compress this down to days."

Onepot AI's service is straightforward: clients select desired compounds from a catalog, and Onepot's technology synthesizes and ships the physical products (as dry compounds or solutions in plates or vials) for the customer's own experiments.

Behind the scenes, Onepot AI's innovation lies in its backend, where Boiko and Tyrin dissect chemical synthesis problems to identify effective molecular combinations. They have built a lab where large language model (LLM) agents access "molecule recipes" for training. These agents learn what works and what doesn't in compound building, generating hypotheses from real-world experimental data rather than relying solely on literature or internet-mined information.

"When executing experiments in the lab, we capture every single detail that goes into the process," Tyrin emphasized, detailing the meticulous tracking of temperature and ingredients. "No information is lost, which makes experiences reproducible even if someone decides to run them in ten years from now."

Funding and Future Outlook

The fundraising process was described as "hectic" by Boiko, who noted that a brief introductory meeting with their lead investor evolved into a multi-hour whiteboard session focused on industrializing synthesis. Beyond Fifty Years, the funding round saw participation from prominent investors including Khosla Ventures, Speedinvest, OpenAI co-founder Wojciech Zaremba, and Google's Chief Scientist Jeff Dean.

The fresh capital will be strategically deployed to establish a second lab in San Francisco, enabling Onepot AI to serve more customers. It will also facilitate the expansion of their team and their compound discovery engine. In the competitive landscape, Onepot AI identifies companies like WuXi AppTec and Enamine as key rivals.

Ultimately, Boiko and Tyrin aspire to make drug discovery at least twice as fast, fundamentally altering perceptions of what's possible in chemistry. They envision unlocking access to previously deemed "off-limits" chemical spaces.

"You're not just speeding up drug discovery, you're expanding the design space for what drugs and materials can be," Boiko concluded. "That drug that we haven't discovered yet, might be out there, waiting for us to find it."