Applied Machine Learning/AI Scientist

Repertoire Immune Medicines•Cambridge, MA

7h•Onsite

About The Position

Repertoire Immune Medicines is a clinical-stage biotechnology company focused on developing transformative therapies for cancer and autoimmune disease by leveraging the power of the human immune system. The company utilizes its proprietary DECODE TM platform, which maps the immune synapse between T cell receptors (TCRs) and their antigen targets, to translate biological insights into targeted immune medicines. Repertoire combines protein engineering expertise with artificial intelligence, utilizing a DECODE database of over one billion TCR-antigen interactions, to accelerate discovery and optimize drug candidates. The company has sites in Cambridge, Massachusetts and Zurich, Switzerland, and is advancing a pipeline of T cell-targeted immunotherapies. Their lead oncology program, RPTR-1-201, is in a Phase 1/2 clinical trial. Repertoire is also developing mRNA tolerizing therapies for autoimmune diseases in partnership with other pharmaceutical companies. Founded in 2019 by Flagship Pioneering, Repertoire has secured four strategic partnerships with major pharmaceutical companies, totaling over $4.5 billion in disclosed deal value.

Requirements

PhD in computational biology, machine learning, engineering, statistics, biostatistics, biomedical engineering, immunology, genetics, cancer biology, or a related quantitative field; or a Master’s degree with 3+ years of relevant industry or academic experience.
Demonstrated ability to deliver impact in cross‑functional, multidisciplinary scientific teams.
Hands‑on experience with protein language models (PLMs), structural modeling, or related ML approaches for biological data.
Familiarity with evaluating and interpreting predicted protein structures, including interface confidence metrics (e.g., pTM, ipTM), and incorporating structural features into machine learning workflows.
Strong programming skills in Python, including experience with scientific and ML libraries such as NumPy, SciPy, pandas, PyTorch, and/or TensorFlow.
Proven ability to analyze and model complex, high‑dimensional biological datasets using sound computational and statistical practices to drive novel insights.
Track record of contributing to scientific publications or equivalent technical outputs (e.g., preprints, conference papers, internal technical reports).
Intellectual curiosity, scientific rigor, and enthusiasm for working in a fast‑paced, evolving research environment.

Nice To Haves

Experience working with TCR-pMHC binding is a strong plus as well as a background in immunology/immuno-oncology.
Practical experience with PLM fine-tuning, embedding extraction, and attention-based interpretation for downstream biological tasks (e.g. binding prediction, fitness landscapes, mutational scanning).
Experience with structural modeling tools and frameworks, including protein structure prediction (AlphaFold2/3, RoseTtaFold, ESMFold), structure-based design (ProteinMPNN, RFdiffusion), and/or graph neural networks operating on 3D protein coordinates (e.g. GCN, raph Transformers, GVP-GNN).

Responsibilities

Assist in the conception, development, optimization, and evaluation of machine learning models to better understand the TCR–peptide‑MHC interface.
Develop, evaluate, and implement rigorous analytical models and methods as needed for scientific discovery and development.
Work alongside other machine learning scientists, computer science engineers, wet-lab scientists, and project managers, contributing to early discovery, lead identification, lead optimization, and biomarker development.
Maintain familiarity with current scientific literature to assist in the development and benchmarking of new methods.
Communicate findings both internally and externally via presentations and publication.