Postdoctoral Scientist - AI & Machine Learning Predictive Drug Absorption

Position: Postdoctoral Scientist - AI & Machine Learning for Predictive Drug Absorption
Shape the Future of Oral Drug Development with AI‑Driven Predictive Science

Last day to apply:
June 30th

Pfizer Research & Development is seeking a highly motivated Postdoctoral Scientist with deep expertise in Artificial Intelligence (AI) and Machine Learning (ML) to advance the prediction of oral drug absorption and formulation performance. In this role, you will play a critical part in developing next-generation predictive models that help transform how drug products are designed, optimized, and translated into clinical success.

You will focus on building, evaluating, and interpreting advanced machine learning models using large, diverse datasets drawn from multiple scientific and clinical sources. Your work will emphasize scalability, interpretability, and real-world applicability ensuring that model outputs are not only technically robust but also scientifically meaningful and decision relevant. A key aspect of this role is the development of explainable modeling approaches, including physics-informed and mechanism-informed learning, to bridge data-driven insights with fundamental pharmaceutical science.

Through this work, you will directly contribute to enabling earlier, faster, and more confident decision-making across Pfizer's R&D portfolio. Your models will help inform formulation strategies, predict in vivo performance, and reduce uncertainty in the development process, ultimately accelerating the delivery of high-quality medicines to patients.

This position is embedded within the Drug Product Design and Supply (DPDS) organization, part of Pfizer's broader Pharmaceutical Sciences division. The role is based in Groton, Connecticut, or Cambridge, Massachusetts, and offers a highly collaborative environment where you will partner closely with interdisciplinary experts across Digital & AI, Clinical Pharmacology, Pharmacometrics, and other quantitative R&D teams. Together, you will integrate cutting-edge AI methodologies with deep domain expertise to solve complex challenges at the intersection of data science and drug development.

Key Responsibilities

* Design, train, and evaluate machine‑learning models for predicting oral drug absorption-related outcomes from high‑dimensional datasets.

* Develop end‑to‑end ML pipelines, including data ingestion, feature engineering, model training, validation, and performance benchmarking.

* Work with large, diverse datasets, including experimental biopharmaceutics data and clinical pharmacokinetic datasets, and internally generated datasets relevant to predictive modelling.

* Apply and compare a range of ML approaches, including tree‑based methods, neural networks, surrogate models, probabilistic approaches for uncertainty‑aware prediction.

* Focus on model interpretability and explainability, linking learned patterns to scientifically meaningful drivers where possible.

* Quantify model robustness, generalizability, and uncertainty, particularly in data‑sparse or extrapolative scenarios.

* Translate ML outputs into actionable insights for drug development teams, rather than purely academic metrics.

* Communicate results through internal technical reports, cross‑functional presentations, and peer‑reviewed publications.

* Contribute to the establishment of AI‑enabled predictive platforms within Pfizer R&D.

REQUIRED QUALIFICATIONS (must have)

* PhD in Machine Learning, Data Science, Applied Mathematics, Computational Sciences, Engineering, Pharmaceutical Sciences, or a closely related quantitative discipline.

* Provide two letters of recommendation with your application (e.g. professors/PI).

* Willingness to commit to the fixed‑term full-time postdoctoral fellowship (duration: 2-4 years).

* Less than 2 years post-doctoral experience.

* At least 1 first-author scientific research article in high-quality specialty or general readership journals.

* Strong foundation in machine learning and statistical modelling, with hands‑on experience building and evaluating predictive models.

* Proficiency in Python and/or R for data analysis and ML development (e.g. scikit‑learn, PyTorch, Tensor Flow, or similar).

* Experience working with large, heterogeneous datasets and structured…