Data Scientist, Chemoinformatics

Atomic AI•South San Francisco, CA

8d•$135,000 - $180,000•Hybrid

About The Position

At Atomic AI, we build artificial intelligence to pioneer new frontiers in drug discovery. Our unique R&D platform, an early version of which was featured on the cover of Science, provides new strategies to treat previously undruggable diseases by targeting RNA. We continue to advance this platform by developing new machine learning methods and unique foundation models fueled by our large-scale, in-house experimental data collection. We are an interdisciplinary team of scientists and engineers and believe our people are our greatest strength and the key to our success. The opportunity As a Data Scientist on the Machine Learning (ML) team, you will extract insights from high throughput datasets for RNA small molecule interactions to advance preclinical drug discovery. You will leverage our technology platform for RNA-targeted drug discovery and drive the development of in silico screening pipelines. Your analyses will inform the development of improved ML models and guide the targeted acquisition of new experimental data. This is a hybrid position with three days in-person at our South San Francisco office.

Requirements

Ph.D. in Computational Chemistry, Chemical Informatics, Chemistry, Biophysics, or a closely related field (or equivalent practical experience).
Strong background in organic or medicinal chemistry and multi-year industry experience applying chemical principles in chemoinformatics workflows.
Strong Python programming skills, including proficiency with cheminformatics-focused libraries such as NumPy, pandas, and RDKit. You’re comfortable moving beyond ad-hoc analyses to develop robust, reusable tools and libraries for internal workflows.
Strong foundation in statistics and experience with conducting statistical analysis of large-scale datasets. You are able to both pose and answer questions about data using statistical methods.
Experience developing, validating, and applying ligand-based modeling approaches, including shape-based methods (e.g., ROCS) and QSAR modeling.
Conceptual understanding of ML model development and evaluation.
Excellent communication and presentation skills, with the ability to clearly and effectively share technical information with colleagues.

Nice To Haves

Understanding of RNA biology and research experience with RNA secondary and tertiary structure.
Background or familiarity with high-throughput experimental assays used for evaluating RNA-SM interactions.
Expertise with structure-based methods for computational drug design and optimization.
Experience developing custom machine learning models (e.g., using scikit-learn) and with deep learning frameworks (such as JAX or PyTorch).
Experience integrating advanced machine learning methods into cheminformatics pipelines.

Responsibilities

Serve as an internal expert on RNA-small molecule (RNA-SM) interactions and small molecule design within the ML team.
Perform statistical analyses to interpret biological variability and extract actionable insights from RNA-SM interaction datasets.
Collaborate closely with the internal wetlab team, influencing the design of experimental assays to probe RNA-SM interactions.
Design and implement computational pipelines for analyzing compound properties in RNA-SM datasets, performing virtual screening, and developing predictive models (e.g., QSAR).
Utilize data analysis and ligand-based modeling to support early-stage drug discovery, including small molecule library design and rational drug design strategies.
Curate small-molecule datasets for training of ML models, help evaluate model performance, and provide directions for improvement.