Mila > AI for humanity > Exascale search of molecules

Exascale search of molecules

About LambdaZero

Drug discovery is lengthy process which costs 2.6B per approved drug on average. The cost is so high because finding a small molecule that binds to a particular target is a perilous highly uncertain process that can’t be completed in a decade in many of the cases. And even when a molecule is found it might not be doing its job very well and might very likely fail in later stages of trials. Our goal is to find much better drug candidate molecules and do it very fast.

What’s Unique About LambdaZero?

Contribute to saving MILLIONS OF LIVES

Play a crucial role in ongoing DRUG DISCOVERY efforts


Collaborate with the foremost experts in AI & MACHINE LEARNING

Have a chance to launch a STARTUP company to scale your idea


Interested in Joining Us?

Join us in developing LambdaZero! We currently have multiple job openings.

View Job Openings

Which Problem Does LambdaZero Aim to Solve?

The number of possible drug-like molecules in the Universe is comparable to the number of atoms in the Universe.

All of the cures found in the history of mankind come from sampling a tiny fraction of this space. The rest of this space likely contains cures for most of the remaining diseases if knew how to search it better.

Our goal is to develop a search technique that could reach the exascale, or 10^18 molecules.

The search in this scale would take 1000 Billion CPU years for existing physics-based methods to simulate.

We define molecules as a set of standard building blocks (think of LEGO bricks) that could be recombined together in various ways. We define search in the molecular space as search in the space of possible configurations of these blocks. Similar to how AlphaZero which uses Monte Carlo Tree Search (MCTS) to master the game of Go, our algorithm called LambdaZero uses MCTS to navigate the space of molecules

LambdaZero’s Performance

Dopamine receptor D4 is an important protein in the human brain. Searching in the larger space of molecules* results in finding molecules with more negative binding energy. More negative binding energy means stronger association between the small molecule and the protein and therefore better binding and better properties as a drug.

*Ultra-large Scale Docking for Discovering New Chemotypes, Lyu at al. Nature, 2019. 

Our Team

Maksym Korablyov


Prateek Gupta


John Bradshaw

Software Engineer

Joanna Chen

Machine Learning Engineer

Bianca Dumitrascu

Machine Learning Engineer

Simon Verret

Machine Learning Engineer

Shenghao Liu

Software Engineer

Cheng-Hao Liu

Machine Learning Engineer

Bruno Rousseau

Machine Learning Engineer

Kostiantyn Lapchevskyi

Software Engineer

Jarrid Rector-Brooks

Machine Learning Engineer

Aga Slowik

Reinforcement Learning Engineer

Emmanuel Bengio

Reinforcement Learning Engineer 

Scott Fujimoto

Reinforcement Learning Engineer

Pierre Thodoroff

Reinforcement Learning Engineer

Clement Gehring

Reinforcement Learning Engineer

Shivam Patel

RL Engineer

Guillaume Alain

RL Engineer

Sacha Leprêtre

HPC Specialist

Frederic Osterrath

Software Engineer

Arnaud Bergeron

Software Engineer

Mike Tyers

Biomedical Assays

Doris Schuetz

Computational Chemist 

Wassim Aouad

Business Development Manager

Lan Dao

Feedback & Culture

Sasha Luccioni

Project Manager

Matt Kusner

Molecule Learning  Advisor

Marvin Segler

Molecule Learning Advisor

Brooks Paige

Molecule Learning Advisor

Will Hamilton

Graph Learning Advisor

Jian Tang

Graph DL Advisor

Michael Bronstein

Representation Learning Advisor

Doina Precup

RL Advisor

Pierre-Luc Bacon

RL Advisor

Charlie Roberts

Clinical Trials Advisor

Yoshua Bengio

Director, Advisor