The YC AI -backed Reactwise is applied to accelerate the manufacture of medicines

Artificial intelligence continues to move things in chemistry. To the orbit: Y Combinator Cambridge, UK -based Reaction Artificial intelligence is used to accelerate chemical manufacturing – an essential step in bringing new drugs to the market.

Once determined by a promising drug in the laboratory, pharmaceutical companies need to be able to produce larger amounts of materials to operate clinical trials. This is the place where Reactwise offers interference with “AI Copilot to improve chemical processes”, which it says is 30x the experiment and the standard error to know the best way to make a drug.

“In the picture above, with the co -founder and CTO Daniel WIGH) in a call with Techcrunch:” drug making is really like cooking, “said co -founder and CEO of Alexander Pomberger. “You need to find the best recipe for making a high purity and high return.”

He said that for years the industry relied on what is summarized in experimental experience and experts or employees for this “development of the process.” Adding automation to the mixture provides a way to reduce the number of repetitions required to drop on a solid prescription for the manufacture of the drug.

The startup believes that it will be able to provide a “single prediction” – where artificial intelligence will be able to “predict the perfect experience” almost immediately, without the need for multiple repetition as data is fed on each experience for more predictions – in the near future (within two years, “Bombarger’s bet).

Automated learning models can still start starting the emerging operation by providing significant savings by reducing the amount of repetition required to bypass this part of the drug development chain.

Boredom

He said, “The inspiration for this is: I am chemical through training, and I worked in Big Pharma, and I saw how the entire industry is boring of experience and error, adding that the work works mainly to unify five years of academic research-to” simple software. “

Reactwise products are “Thousands” of the reactions made by starting up in their laboratories in order to capture data points to feed their AI’s predictions. Pomberger says that startup used the “high productivity examination” method in its laboratory, which allowed her to examine 300 reactions at one time, allowing it to accelerate the capture process of all these training data for artificial intelligence.

“In Pharma … there are one or two reactions, and the types of reaction, which are used again and again,” he said. “What we do is that we have a laboratory where we create thousands of data points for these most relevant reactions, and training the constitutive reaction models on our side, and these chemistry models can be understood mainly.

The startup started to capture the reaction types of AIS training last August, and Pomberger said it would be completed by summer. It works for an extension of 20,000 chemical data points “to cover the most important reactions”.

He said: “To get one data point in a traditional way, it is a pharmacist, usually from day to three days,” he said, adding: “Therefore, we call it expensive to evaluate the data.

It focuses so far on manufacturing operations for “small molecules”, which Pomberger said can be used in medicines that target all types of diseases. But he suggested that this technology can be applied in other disciplines, noting that the company is also working with two material manufacturers in the development of a pharmaceutical development.

RECTWISE Autism Operation also includes programs that can interact with the automated laboratory equipment to connect the manufacture of exact medications. Although it is clear, it focuses purely on selling programs; She is not a robotic laboratory group maker itself. Instead, it adds another chain to the bow in the ability to provide the payment of automated laboratory equipment if its customers have a group of this group.

An emerging company in the United Kingdom, which was established in July 2024, has 12 experimental experiments for its program and its operation with pharmaceutical companies. Pomberger said they expect the first transfers-to the widely publishing operations of the subscription program-later this year. Although he has not yet revealed the names of all companies that work with them, Reactwise includes some of the adult Parma players.

Funding before a seed

Reactwise reveals the full details of the pre -seed lifting, which amounted to a total of $ 3.4 million, Startup Techcrunch exclusively told.

The number includes a pre -unveiled support from YC ($ 500,000) and Creating a UK grant From approximately 1.2 million pounds (about $ 1.6 million). The rest of the financing (about $ 1.5 million) comes from investment capital and owners who say, who say “the commitment to progress in sustainable and sustainable pharmaceutical industrialization.”

While Reactwise focuses, somewhat narrowly, on a specific part of the drug development chain, Pomberger said that the acceleration here can make a meaningful difference in reducing the time it takes to obtain new drugs for patients.

“Let’s take a look at the duration of a property from start to 10 to 12 years.

At one time, other startups Application of artificial intelligence on various aspects of the development of the drugIncluding the identification of interesting chemicals in the first place, so there are likely to have effects with folding more automation innovations.

But when it comes to manufacturing medications, specifically, Pomberger argues that Reactwise is applying for the package. “We were the first to actually treat this,” he said.

The start starting with old programs is competing with statistical methods, such as JMP. He also said that there are a few others who apply artificial intelligence to accelerate the manufacture of medicines, but he said that Reactwise’s access to high -quality data groups on chemical reactions gives them the competitive edge.

He said: “We are the only ones who have the ability to be able, and who are currently being generated, high -quality data sets at home.” “Most of our competitors provide the program.

“But on our part of things, we offer these pre-models-and you have a very strong because they mainly understand chemistry.