Senior Scientist, Generative AI & Agentic Modeling

By clicking the “Apply” button, I understand that my employment application process with Takeda will commence and that the information I provide in my application will be processed in line with Takeda’s Privacy Notice and Terms of Use. I further attest that all information I submit in my employment application is true to the best of my knowledge.

Job Description

Position Overview

At Takeda, we are building the next generation of intelligent R&D platforms by integrating Generative AI and agentic AI systems into drug discovery. We seek a Senior Scientist with deep expertise in AI model development to lead efforts in designing foundation models and autonomous agents that drive innovation across our therapeutic areas, including oncology, neuroscience, and inflammatory diseases.

This execution-focused role centers on the development and deployment of LLMs, generative protein models, and autonomous reasoning agents within an AI-first computational biology group. The ideal candidate will be a builder — someone who can translate cutting-edge research in generative modeling into scalable, drug-discovery tools and autonomous pipelines.

Key Responsibilities

• Design and implement next-generation generative AI models for biologics discovery, including LLMs, protein language models, diffusion models, and multi-modal architectures.
• Build and refine agentic AI frameworks (e.g., tool-using agents, retrieval-augmented generation (RAG), memory-augmented LLMs) to autonomously plan and optimize design–make–test–analyze (DMTA) cycles.
• Collaborate with bioinformatics, automation, and wet-lab teams to build lab-in-the-loop systems, enabling AI agents to initiate hypotheses, request experiments, and iterate based on outcomes.
• Develop reusable AI pipelines that integrate multi-objective optimization (e.g., binding affinity, immunogenicity, developability, manufacturability) from in vitro and in silico data sources.
• Lead or contribute to the development of internal foundation models for biological design, including fine-tuning of LLMs on proprietary sequence, omics, and experimental datasets.
• Prototype and deploy agentic AI solutions using industry-standard frameworks such as LangChain, OpenAI API, Hugging Face Transformers, or similar.
• Stay current with advancements in transformer architectures, RLHF, memory systems, cognitive planning agents, and apply these innovations to real-world therapeutic discovery challenges.
• Clearly communicate ideas to cross-functional stakeholders and contribute to internal and external scientific knowledge-sharing.

Required Qualifications

• PhD degree in Computer Science, Computational Biology, Machine Learning, Bioinformatics, or related field (or equivalent) with 2+ years relevant experience, or MS with 8+ years relevant experience, or BS with 10+ years relevant experience Demonstrated expertise in developing large-scale deep learning models, including LLMs, generative models (e.g., diffusion, VAEs, autoregressive transformers), or multi-modal architectures.
• Proficiency in Python and modern ML frameworks (e.g., PyTorch, Hugging Face, JAX, LangChain); strong software engineering and reproducibility skills.
• Experience applying machine learning to biological sequences, experimental data, or computational biology pipelines (e.g., protein or antibody engineering, `omics analysis, etc.).
• Experience working with large datasets (NGS, high-throughput screens, bioassays) for supervised or unsupervised model training and evaluation.
• Deep familiarity with agentic AI design: planning agents, tool-using agents, memory-augmented LLMs, or lab automation interfaces (e.g., lab robots, simulation frameworks, or autonomous workflows).
• Proven ability to integrate AI models into production environments and iterative workflows, preferably within a pharmaceutical or biotech con