Harmonic Discovery

We engineer drugs with targeted polypharmacology

Senior Scientist, Machine Learning for Drug Discovery at Harmonic Discovery

New York City, Helsinki, Anywhere / Remote
Job Type
1+ years
Connect directly with founders of the best YC-funded startups.
Apply to role ›
Jason Lee
Jason Lee

About the role

About Harmonic Discovery

Based in NYC, Harmonic Discovery is a venture-backed biotechnology company using machine learning, biology, and medicinal chemistry to create a new generation of multi-target therapeutics for oncology and autoimmune disorders. At Harmonic Discovery, we are building a multifaceted and talented team of machine learning engineers, biologists, and chemists.

We look for passionate innovators, radical thinkers and collaborative builders. It’s time to integrate our understanding of biology and chemistry with better AI to make superior cures that make a difference in people’s lives.

Harmonic Discovery is backed by top tier venture firms and is a member of Johnson & Johnson’s accelerator – JLABS @ NYC. We are also backed by YCombinator.


Harmonic Discovery is looking for a Machine Learning Scientist to lead our computational chemistry/drug discovery platform. We are recruiting exceptional scientists to

• Lead the development of deep learning models for chemical structure featurization and de novo compound generation;

• Help recruit and build our machine learning and computational chemistry team;

• Work on compound and protein feature engineering;

• Create and implement novel, biologically-inspired, machine learning architectures for the prediction of drug-protein interactions.

Harmonic Discovery is committed to a multicultural environment and strongly encourages applications from women and underrepresented minorities.


• PhD in Machine Learning, Statistics, or other STEM fields. Exceptional MS and BS candidates will be considered. Experience in computational chemistry is a plus.

• Demonstrated experience in developing machine learning models, especially leveraging deep learning approaches

• Strong knowledge of mathematics, statistics and computer science.

• Enthusiastic about application of computational methods to drug discovery.

• Strong communication skills.

• Collaborative, team-oriented mindset.


• Strong Python programming skills.

• Experience with Python libraries such as scikitlearn, matplotlib, numpy, etc.

• Experience with deep learning platforms, such as TensorFlow, H2Oo.ai, and PyTtorch.

• Experience in bash and shell scripting.

• Familiarity with AWS computing services.


• Experience with generative machine learning models (variational autoencoders, RNNs).

• Experience with graph-based modeling (GCNNs).

• Experience with training deep models on either biological or chemical datasets.


• Highly competitive salary, benefits (healthcare, 401k matching, etc.) and stock options.

• Significant responsibilities with potential for growth.

• Work from anywhere! Come by the office if you feel like it, or don’t. Your call. Feel free to work from anywhere in the US if you have a stable internet connection and can synchronize your calendar’s time zone with the team’s.

• You will work closely with the entire team, from the scientists to the CEO.

• No busy work here, you’ll work on projects that have an immediate and obvious impact to the company.

• A great culture. We are building a flexible and friendly work environment in the lab or at home.

Harmonic Discovery is an equal opportunity employer. Harmonic Discovery recruits and employs regardless of race, religion, color, national origin, sex, disability, age, veteran status, and other protected status as required by applicable law.

Why you should join Harmonic Discovery

We are a drug discovery company developing a new generation of therapeutics that embrace the complexity of disease. Currently available approaches for creating drugs work on the principal of finding one drug-one protein 'magic bullets'. However, diseases such as cancer and autoimmunity are often the result of several dysregulated proteins across many distinct biological pathways.

We are building a computational-experimental platform to design therapeutics that can target several disease-causing proteins at once. By designing multi-specific drugs, we are able to create therapeutics that are more efficacious and safer than existing medicines.