AI Engineer – Discovery

AstraZeneca•Cambridge, MA

1d•$124,000 - $187,000

About The Position

EsoBiotec, now part of AstraZeneca, is seeking a cross-disciplinary AI Engineer to support a wet-lab-focused discovery research team working in CAR-T, in vivo lentiviral delivery, and oncology. The role applies AI, machine learning, and data-centric engineering to experimental workflows, biological data, and decision-making across early discovery and platform development. The successful candidate will partner closely with bench scientists to understand biological questions, experimental systems, and data-generation challenges, then translate them into practical computational approaches that improve insight generation, prioritization, and research efficiency. The work focuses on cell therapy, gene delivery, oncology discovery, screening, vector design, cellular profiling, functional assays, and iterative platform optimization. It is not primarily focused on medicinal chemistry or small-molecule discovery. The position involves building fit-for-purpose models and tools, connecting experimental output with computational decision support, and collaborating across scientific and technical teams. The annual base pay ranges from $124,000 to $187,000. The position may also be eligible for short-term incentive bonuses and equity-based awards for salaried roles. Benefits include qualified retirement programs, paid time off, and health, dental, and vision coverage, subject to applicable plan terms.

Requirements

Bachelor’s degree in computer science, machine learning, computational biology, bioinformatics, biomedical engineering, systems biology, applied mathematics, or a related field.
At least 5 years of experience developing AI/ML solutions in biological, biomedical, or R&D environments.
Strong programming skills in Python and experience with relevant data science and machine learning frameworks.
Experience working with complex experimental datasets and building practical analyses or tools that support scientific decision-making.
Ability to work effectively with wet-lab researchers and communicate clearly across computational and experimental disciplines.
Ability to handle ambiguity and translate open-ended scientific questions into structured computational approaches.
Strong grounding in data quality, model evaluation, reproducibility, and analytical rigor.

Nice To Haves

An advanced degree, such as a Master’s or PhD.
Experience in cell therapy, CAR-T, gene therapy, viral vector biology, lentiviral systems, immuno-oncology, or functional genomics.
Familiarity with relevant biological data modalities, including flow cytometry, single-cell RNA-seq, multimodal omics, imaging, perturbation datasets, and high-dimensional assay data.
Experience with multimodal learning, representation learning, scientific image analysis, active learning, Bayesian optimization, LLM-enabled scientific workflows, or knowledge graph approaches.
Understanding of discovery-stage wet-lab research realities, including assay variability, small and noisy datasets, metadata challenges, and iterative hypothesis testing.
Experience developing user-facing tools, dashboards, or decision-support applications for scientific teams.
Familiarity with cloud, data engineering, or production-oriented workflows for research computing environments.

Responsibilities

Partner with experimental scientists in CAR-T, lentiviral delivery, cell engineering, and oncology discovery to understand scientific questions, assay workflows, and experimental decision points.
Translate wet-lab problems into AI and data science opportunities, including construct prioritization, delivery optimization, cell-state characterization, image-based phenotyping, multimodal data integration, and prediction of experimental outcomes.
Build, adapt, and apply AI/ML models to support discovery, including predictive and classification models, clustering, representation learning, computer vision pipelines, and LLM-enabled knowledge workflows where appropriate.
Work with biological datasets such as flow cytometry, single-cell and bulk omics, imaging, functional assay readouts, vector characterization data, metadata, and experimental annotations.
Develop data pipelines and analytical frameworks that improve data quality, accessibility, interoperability, and reuse across experimental programs.
Create practical tools and visualizations that help bench scientists explore data, compare constructs or conditions, and decide what to test next.
Collaborate with biologists, immunologists, gene therapy scientists, bioinformaticians, data scientists, and software and informatics partners.
Evaluate and communicate model performance with attention to biological validity, robustness, interpretability, and limitations in real experimental settings.
Help establish best practices for reproducible, responsible, and scientifically grounded AI use in discovery research.
Monitor emerging methods in AI, computational biology, and multimodal learning and identify meaningful applications in CAR-T and in vivo delivery research.