Atomwise is a leading artificial intelligence (AI) drug discovery company, based in San Francisco, CA. We discover and develop small molecules that will improve human health and agricultural productivity.
Our team has over 50 Ph.D. scientists who contribute to a collaborative academic-like culture that fosters robust scientific and technical discussion. We strongly believe that data wins over opinions, and aim for as little dogma as possible in our decision-making. Our team members have expertise in a wide range of disciplines–from computational chemistry and structural biology to cloud-native best practices–and we regularly have internal seminars open to anyone interested in learning about these topics.
Our ml.research team is responsible for developing new tools and techniques to be used by our applied scientists. We are looking for a Senior Machine Learning Scientist to improve our abilities in molecular property prediction and optimization for lead optimization. As a member of our team, you will:
- Design, implement, and evaluate state-of-the-art algorithms in molecular optimization and design in a potency and ADMET prediction setting.
- Improve our ligand-based modeling techniques with the goal of improving out-of-distribution performance when limited training data is available.
- Develop generative models for molecular optimization.
- Improve the interpretability of our models, and the confidence bounds on their predictions.
- Engage actively in team collaborations, clearly and efficiently report and present research findings and developments, write academic papers on novel developments.
Our ml.research team is small and growing quickly. As a result, there are plenty of opportunities to have a big impact on our success.
- PhD. or M.Sc. in Machine Learning, Computer Science, Computational Chemistry, Statistics, or related fields.
- 5+ years of relevant work experience in machine learning guided molecular optimization.
- Hands-on experience with model design/development, training, validation, and deployment.
- Advanced knowledge with Python-based scientific computing and data analysis: NumPy, scikit-learn, SciPy, Pandas, PyTorch (or TensorFlow), and Matplotlib.
- Strong analytical and statistical skills.
- Scientific rigor, healthy skepticism, and detail-orientation in training and analyzing machine learning models.
- Familiarity with modern research techniques in molecular optimization and/or molecular property prediction on ADMET modeling, protein-ligand interactions, or related problems, such as:
- Self-supervised learning and representation learning,
- Few-shot and one-shot learning,
- Active learning,
- Generative models (GANs, VAEs, normalizing flows) or Reinforcement Learning,
- Out-of-distribution generalization of predictive models in constrained data settings,
- Experience designing and building modern neural networks for learning on molecules (Graph Neural Networks, Recurrent Neural Networks, Transformers).
- Experience with best software development practices in an agile and collaborative environment.
- Excellent problem-framing, problem-solving and project management skills
- Ability to work independently.
- A real passion for Machine Learning!
Compensation & Benefits
- Great, world-class team of colleagues – scientists from a variety of backgrounds (chemistry, medicine, biology, physics, CS/ML)
- Stock compensation plan – you’ll be an Atomwise co-owner
- Platinum health, dental, and vision benefits
- 401k with 4% match
- Funding for professional development and conference attendance
- Flexible work schedule
- Generous parental leave
Atomwise is an equal opportunity employer and strives to foster an inclusive workplace. Our mission is to develop better medicines faster, and we know that we need a diverse team to develop medicines that serve diverse populations. Accordingly, Atomwise does not make any employment decisions (including but not limited to, hiring, compensation, and promotions) on the basis of race, religion, color, national origin, gender, gender identity, sexual orientation, age, veteran status, disability status, or any other characteristics protected by applicable federal, state, and local law.
We strongly encourage people of diverse backgrounds and perspectives to apply. Authorization to work in the U.S is required. Pursuant to the San Francisco Fair Chance Ordinance, we will consider for employment qualified applicants with arrest and conviction records.
Application instructions can be found at this LINK.