Takeda is seeking to accelerate the next generation of biologics discovery by pioneering AI-driven platforms and integrating state-of-the-art AI with experimental strategies to drive therapeutic breakthroughs in oncology, neuroscience, and inflammatory diseases.
Requirements
- Demonstrated ability to build and apply ML models (deep learning, protein LMs, GNNs) to biological or structural datasets.
- Hands-on experience with antibody design or protein engineering, especially using ML-guided or structure-aware methods.
- Strong understanding of protein structure and dynamics, including MD simulation, FEP, Rosetta modeling, or AlphaFold-based tools.
- Proficiency in Python and ML libraries (e.g., PyTorch, scikit-learn, NumPy); familiarity with Unix command line tools and scripting.
- Strong data wrangling and visualization skills; ability to translate modeling outputs into actionable insights for bench scientists.
- Experience developing or fine-tuning generative models for protein design (e.g., RFdiffusion, ProGen, ESM-IF, ProteinMPNN).
- Prior application of structure-guided ML to engineer antibodies, antigens, or binders against defined epitopes.
Responsibilities
- Develop and implement AI/ML models for sequence- and structure-based design of biologics, with emphasis on generative frameworks (e.g., diffusion models, inverse folding, LMs).
- Apply and extend tools such as Boltz, Rosetta, AlphaFold2, ESMFold, ProteinMPNN, and other foundational models for de novo antibody discovery, affinity maturation, and developability optimization.
- Build and optimize predictive models for multiple objectives (e.g., solubility, immunogenicity, thermostability, epitope specificity) based on NGS, in vitro, and in vivo datasets.
- Integrate 3D structural data into model-guided protein design pipelines.
- Collaborate with experimental scientists to inform hypothesis generation, model validation, and iterative learning in Design–Make–Test–Analyze cycles.
- Manage and process large-scale experimental and synthetic datasets for model training, benchmarking, and deployment.
- Prototype and deploy ML pipelines using best practices in software engineering and reproducibility
Other
- PhD degree in Computational Biology, Structural Biology, Machine Learning, or related fields (or equivalent) with 2+ years relevant experience, or MS with 8+ years relevant experience, or BS with 10+ years relevant experience
- Excellent interpersonal and written communication skills; thrives in a highly collaborative environment.
- Clearly communicate complex ideas to technical and non-technical audiences and contribute to internal knowledge sharing.
- Exposure to wet-lab data integration and cross-functional collaboration with experimental biologists.
- Understanding of biologic drug properties (developability, immunogenicity, CMC constraints).
