Data Scientist II, ML Expert

AbbVie•North Chicago, IL

1d•$84,500 - $162,000

About The Position

Conceive and implement novel experimental approaches to answer scientific questions. Investigate, identify, develop, and optimize new methods and techniques. Contribute to project science in area of expertise. Develop productive collaborations and communications with other groups across science disciplines.

Requirements

Bachelor’s Degree or equivalent education and typically 7+ years of experience, Master’s Degree or equivalent education and typically 5+ years of experience.
Degree in relevant field (chemistry / computer science / machine learning / cheminformatics / chemical engineering) with experience developing machine learning models related to chemical and biological data, preferred.
Possess thorough theoretical and practical understanding of own scientific discipline.
Expertise with developing, implementing and deploying programs and computational solutions employing Machine Learning/Deep learning and Cheminformatics.
Expertise in AI/ML-enabled molecular generation, pose prediction, affinity prediction and/or prediction of pharmacological properties (e.g. PK).
Strong programming skills in Python and experience with data science stack including numpy, pandas, scikit-learn, and other related scientific libraries.
Ability to implement, debug, and maintain computational tools in common programming languages (Python, etc…) and proficiency with cloud computing capabilities.
Familiarity with modern deep learning architectures including GNN, CNN, RNN, Transformer, GCNN and MPNN, and machine learning paradigms such as generative models, GAN, and active learning.
Strong analytical and problem-solving skills with demonstrated ability to think critically and creatively, and provide solutions both individually and collaboratively with internal experts.
Excellent ability to communicate clearly and concisely with colleagues and collaborators including an ability to explain complex ideas to non-specialists.

Responsibilities

Consult with project team to establish quantitative targets for driving optimization
Work with project team to identify suitable models to predict small molecule binding affinities for protein targets, and to predict pharmacological properties (PK parameters, etc.)
Work with project team to build predictive models for each relevant endpoint, and carefully benchmark to evaluate performance and domain of applicability for each model
Work with project team to identify the model that should be used for each project team query molecule, using data-driven approaches that take account of model uncertainty and heteroscedasticity
Clearly communicate to the project team the rationale for model selection and why the selected models are best suited for the task at hand