Job Description

Locations

• India - Hyderabad

Skills Required

• modern deep learning frameworks such as PyTorchintermediate
• Pythonintermediate
• Python scientific computing toolsintermediate
• foundation modelsintermediate
• reproducible machine learningintermediate

Required Qualifications

• Doctorate degree with 4+yrs in Data Science, Computer Science, Computational Biology, Bioinformatics, Computational Chemistry, or a related field (degree in data science)
• Master's degree and 8+ years of directly related experience (experience, 8 years)

Preferred Qualifications

• Experience developing machine learning models for biological, protein engineering, drug discovery or immunology applications (experience)
• Strong proficiency in Python and experience with modern deep learning frameworks such as PyTorch (experience)
• Experience with Python scientific computing tools, such as numpy, scipy, pandas, etc. (experience)
• Experience with foundation models, representation learning, sequence-to-property modeling, and multimodal learning (experience)
• Experience integrating sequence, structural, assay, imaging, and metadata sources into reliable, model-ready datasets (experience)
• Proven ability to apply active learning, Bayesian optimization, uncertainty quantification and model interpretation to scientific problems (experience)
• Experience working with large-scale datasets and in high-performance computing environments, including cloud-based platforms (e.g., AWS) (experience)
• Familiarity with reproducible machine learning and software engineering practices, including version control (git), testing, containerization (Docker), workflow orchestration, data versioning, or experiment tracking. (experience)
• Strong scientific communication skills, with publications in leading ML, computational biology, or protein science venues such as NeurIPS, ICML, ICLR, ISMB, Nature Biotechnology, Nature Methods, Cell Systems, MABS, PNAS or comparable conferences and journals; candidates should highlight representative publications on their resume. (experience)
• Experience developing machine learning models for biological, protein engineering, drug discovery or immunology applications (experience)
• Strong proficiency in Python and experience with modern deep learning frameworks such as PyTorch (experience)
• Experience with Python scientific computing tools, such as numpy, scipy, pandas, etc. (experience)
• Experience with foundation models, representation learning, sequence-to-property modeling, and multimodal learning (experience)
• Experience integrating sequence, structural, assay, imaging, and metadata sources into reliable, model-ready datasets (experience)
• Proven ability to apply active learning, Bayesian optimization, uncertainty quantification and model interpretation to scientific problems (experience)
• Experience working with large-scale datasets and in high-performance computing environments, including cloud-based platforms (e.g., AWS) (experience)
• Familiarity with reproducible machine learning and software engineering practices, including version control (git), testing, containerization (Docker), workflow orchestration, data versioning, or experiment tracking. (experience)
• Strong scientific communication skills, with publications in leading ML, computational biology, or protein science venues such as NeurIPS, ICML, ICLR, ISMB, Nature Biotechnology, Nature Methods, Cell Systems, MABS, PNAS or comparable conferences and journals; candidates should highlight representative publications on their resume. (experience)

Responsibilities

• Let's do this. Let's change the world. We are seeking a Senior Data Scientist with strong expertise in machine learning and predictive modeling to join the AI & Data for Engineered Biologics team within Amgen's Large Molecule Discovery (LMD) organization. In this vital role you will enable data-driven design and optimization of biologics by developing models that predict key developability and therapeutic properties. You will work with sequence, structure, and experimental datasets to generate actionable insights and guide discovery decisions.
• The successful candidate will work in a highly collaborative, multidisciplinary environment, partnering with different experimental teams to build predictive models, train and evaluate protein foundation models, and design active-learning strategies that improve data generation and candidate optimization.
• Develop and apply predictive machine learning models for biologics properties, including developability and clinical immunogenicity-related endpoints
• Train, fine-tune, and evaluate foundation models for sequence/structure-to-property prediction and biologics design
• Integrate multimodal data, including protein sequence, structure, imaging, assay, and experimental metadata, to improve predictive performance and model robustness
• Design active learning, uncertainty-aware, or Bayesian optimization strategies to prioritize experiments and guide data generation across modalities
• Build reproducible modeling workflows and evaluation frameworks that support rigorous model comparison, validation, and deployment-readiness
• Collaborate cross-functionally with scientists and software engineers to translate machine learning innovation into actionable insights for biologics discovery and optimization
• Communicate scientific findings and modeling strategies through presentations, technical documentation, and cross-functional forums
• Develop and apply predictive machine learning models for biologics properties, including developability and clinical immunogenicity-related endpoints
• Train, fine-tune, and evaluate foundation models for sequence/structure-to-property prediction and biologics design
• Integrate multimodal data, including protein sequence, structure, imaging, assay, and experimental metadata, to improve predictive performance and model robustness
• Design active learning, uncertainty-aware, or Bayesian optimization strategies to prioritize experiments and guide data generation across modalities
• Build reproducible modeling workflows and evaluation frameworks that support rigorous model comparison, validation, and deployment-readiness
• Collaborate cross-functionally with scientists and software engineers to translate machine learning innovation into actionable insights for biologics discovery and optimization
• Communicate scientific findings and modeling strategies through presentations, technical documentation, and cross-functional forums

Tags & Categories