Yoneda Labs
Foundation Model for Chemical Manufacturing
Yoneda Labs provides software to help chemists optimise reactions. When chemists make a drug or a material, we help them figure out the best reaction parameters such as temperature, concentration and catalyst. When Jan was working at chemical labs, he experienced the struggle of spending weeks guessing reaction conditions. We then started experimenting with ML to speed up the process. Now, as a team of three friends from the University of Cambridge, we’ve spent the last month combining our domain expertise in Computer Science, Machine Learning and Chemistry to develop state of the art models for reaction optimisation. Although ML is becoming well established in other fields, current chemical models generalise poorly and require lots of programming experience. We make our models easily accessible to chemists in the lab. Finding the right conditions quickly allows pharmaceutical companies to test more drugs, and finding better optima makes manufacturing process cheaper and more environmentally friendly.
Latest News
Jobs at Yoneda Labs
Active Founders
Michał Mgeładze-Arciuch, Founder
Michał is the Co-Founder and CEO of Yoneda Labs. Michał studied Computer Science at the University of Cambridge. He has strong software engineering and Artificial Intelligence experience having worked at Jane Street and Berkeley AI Research.
Jan Oboril, Founder
Jan is the Co-Founder and Chief Scientist at Yoneda Labs. Jan studied Chemistry & Biology at the University of Cambridge. Has worked at Bayer and as a research chemist for institutes in Czechia and Austria. He has a lot of experience designing ML models to solve problems in Chemistry and Biology, most notably he designed and sold a model for cell segmentation. He scored 8th/10th/11th at the international Chemistry Olympiad.
Daniel Vlasits, Founder
Daniel is the Co-Founder and CTO at Yoneda Labs Daniel studied Computer Science at the University of Cambridge. He has strong software engineering experience having been a software engineer at Cisco, Optiver and a programming languages researcher at the University of Cambridge. He won first place at the international robotics competition Pi Wars, and ranked 5th in the Cambridge Computer Science cohort.
Company Launches
TL;DR
Yoneda Labs is building the foundation model for chemical reactions. When chemists need to make a drug we use AI to define parameters such as temperature, concentration and catalyst. This enables new chemistries and makes synthesis faster and cheaper.
Check out our product here: www.yonedalabs.com/yoneda-optimize
Who we are
We are three University of Cambridge grads combining expertise in Computer Science and Chemistry, with a leader in the field, Prof Scott E Denmark as our advisor. Combined, we have experience working in Bayer, Berkeley AI Research, Jane Street, Optiver and Cisco. Jan ranked top 11 in the International Chemistry Olympiad for 3 years in a row.
Problem: There are more than 10M different compounds created each year. Chemists need to figure out how to make them.
To do so, they often rely on literature data. Unfortunately, this data almost never contains the exact recipe and the ones out there are often misreported and irreproducible in the lab. This means that chemists still need to frequently rely on their intuition, and when that fails, a potentially life-saving drug might never see the light of day.
Solution: Foundation model for chemical reactions
We are generating our own dataset of chemical experiments. We will use it to train a powerful model capable of defining good conditions for any organic reaction.
We’ve performed feasibility studies on smaller datasets and confirmed our beliefs that if we choose the experiments very carefully, we will be able to cover 3 of the most popular reactions in organic chemistry with only 20k data points.
We’ve already created an app that learns from the chemists’ experimental data as they run their reactions in the lab. This allows them to find the optimal conditions in days instead of weeks or months.
Our custom algorithms allow the app to run fully offline from the chemist's PC to account for data sensitivity. Additionally, we designed the app to be simple to use so that every chemist can employ state-of-the-art statistics and machine learning in their lab in just a few clicks. You can check out what it looks like here.
Our Ask
- If your company works with chemical processes, we’d love for you to try our product!
- Any intros to chemistry CROs / pharma companies especially key decision makers responsible for the medicinal and process chemistry departments.
Contact us here: michal@yonedalabs.com
Book a demo here