AIRx Director, Computational & AI Biologics Design Lead

Job Description

Director, Computational & AI Biologics Design Lead

Takeda Research is constructing a Lab of Tomorrow built on AI, automation, new ways of working, and talent with the singular vision of delivering differentiated medicines to the clinic at speed and cost. To catalyze these efforts, Takeda is creating two complementary units: AI Research Accelerator (AIRx) and Discovery Automation & Robotics (DAR).

AIRx will have a dedicated group of experienced biologic drug hunters with the autonomy of a biotech and the resources of a leading pharmaceutical company. It is designed to incubate the future AI-powered operating models for large molecule discovery and deliver candidates to the clinic at industry leading speed and success rates.

Reporting to the Head of AIRx, the Computational & AI Biologics Design Lead sits at the scientific heart of the Takeda Boston (TBOS) Large Molecule Pod. As part of the AIRx team, this role serves as a strategic computational leader, driving in silico biologics design and shaping how AI-enabled biologics discovery is executed for select programs. This role drives in silico biologics design, applies generative AI and structure-informed methods to antibody and large-molecule programs, and connects Takeda’s internal AI/ML platform capabilities to the day-to-day decisions of a fast-moving drug-hunting team.

In addition, the role defines decision frameworks and scientific standards that influence candidate prioritization, progression, and overall portfolio direction. Proposals from this role directly shape what gets engineered, what gets deprioritized, and what ultimately reaches the clinic. The role is deeply hands‑on, with a mandate to operate with urgency and independence while remaining tightly integrated with biology, protein engineering, and translational science colleagues across the pod.

1. In Silico Biologics Design for Pod Programs

• Define and drive the computational design strategy across the pod’s large-molecule programs, including antibody, VHH, and multispecific or fusion formats, from early format selection through lead optimization.
• Design and prioritize molecular candidates using generative AI/ML and computational modeling approaches.
• Serve as a scientific advisor to pod leadership on computational design decisions, influencing program direction and key trade-offs.
• Partner closely with the Biologics Discovery Lead to translate computational proposals into testable engineering priorities; challenge and be challenged on scientific assumptions in equal measure.
• Integrate structural biology data into design strategies to inform format selection, epitope targeting, and interface optimization.
• Oversee virtual screening, binding affinity prediction, and develop ability risk assessment for candidate sequences; provide ranked shortlists with quantified uncertainty to the pod.
• Establish and improve approaches to accelerate lead optimization by compressing DMTA cycles through AI-guided design, with the goal of achieving the target candidate profile in fewer rounds.
• Collaborate with translational and DMPK scientists to model PK/PD behavior, TMDD, and species cross-reactivity in silico, informing study design and reducing in vivo cycle time.

2. AI/ML Platform Interface and Data Strategy

• Serve as the pod’s primary computational interface to Takeda’s AI/ML research platform; evaluate and benchmark new AI design tools against the pod’s specific biologics modalities and program needs.
• Define and steward data requirements for AI model training within the pod: structure data return from experimental campaigns, annotation standards, and integration with Takeda’s data infrastructure.
• Contribute to building and curating AI/ML training datasets from pod experimental outputs to enable continuous model improvement.
• Guide development and refinement of computational workflows to enable pod scalability, speed and reproducibility across the DMTA cycle; document methods to support cross-pod learning.

3. Pod Integration and Scientific Operations

• Act as a hands-on computational authority within pod governance: prepare and present in silico analyses for PRC reviews,…