Principal (Senior Director), AI Product Management (Portfolio Systems)

Position Summary

GSK's Enterprise AI team exists to translate the promise of AI into measurable impact across every major group at GSK — R&D, Medical, Commercial, Supply Chain, HR, Legal, Finance, and beyond. You will act as the strategic compass for driving a portfolio of AI solutions that transform how GSK runs its technology and R&D portfolios — from intake, prioritization, and funding stages through resource allocation, benefits tracking, and portfolio reviews — and own the forward deployed strategy each pod executes against.

This is a senior individual contributor role for someone who can see across GSK's value chain, identify where AI will create disproportionate value, prototype hands on, and ensure engineers are always deployed against the highest-impact opportunities. You will not manage engineers; you will shape what they build, where they go next, and how we measure whether it worked.

Responsibilities

• Hunt for high-value AI opportunities across the portfolio management and value-realization lifecycle — intake, prioritization, funding stages, resource allocation, benefits tracking, and portfolio reviews — and develop a global portfolio of AI products ranked by impact, feasibility, and time-to-value.
• Quantify value rigorously by building business cases that hold up with finance and executive sponsors — cycle time saved, decisions accelerated, cost avoided, and capacity returned to portfolio and delivery teams.
• Develop domain pattern recognition in portfolio management workflows so you can spot AI-shaped problems the business itself hasn't yet articulated.
• Own the Enterprise AI opportunity portfolio, maintaining a living view of in-flight, queued, and de-prioritized opportunities, with the rationale for each decision.
• Make hard trade-off calls — building the framework, defining success metrics, running the analysis, and recommending where engineers go when business units compete for the same resources.
• Say no, with evidence, to opportunities that look attractive but won't move scalable outcomes, protecting engineering focus.
• Be the strategic counterpart to the forward deployed, full-stack, and data engineering leads, translating ambiguous business problems into crisp problem statements, prototypes, success metrics, and decision frameworks engineers can build against.
• Shape engagements end-to-end — defining scope, success criteria, exit conditions, and scaling pathways before kick-off, then staying close as engineers execute.
• Surface re-use and platform leverage, spotting when a one-off solution should become a reusable capability and working with engineering to make that case.
• Design the human-digital split, deciding which tasks the pod's digital workers own and where engineers and business users stay in the loop.
• Redesign the portfolio operating model around automation, defining decision rights, gates, cadences, roles, and how work moves between people and digital workers.
• Build a reusable skills library, establishing shared agent skills, prompts, templates, and connectors that multiple teams contribute to and reuse.
• Instrument real-time portfolio insight, bringing together delivery and value data from engineering tools, financial systems, usage telemetry, and AI platforms.
• Operate fluently with senior business and scientific leaders, running executive conversations that turn fuzzy ambitions into concrete, fundable AI engagements.
• Build sponsor coalitions, securing the air cover, data access, and change-management commitment that engineering deployments require to succeed.
• Communicate impact clearly, telling the value story to ELT and the board, in language that connects AI work back to patients, science, and shareholder outcomes.
• Define how we keep score, establishing the value-realization framework Enterprise AI uses to measure impact across deployments.
• Run honest post-mortems — capturing what worked, what didn't, and feeding those learnings back into opportunity selection and engagement design.
• Continuously sharpen the investment thesis on where AI most belongs at GSK, and where it doesn't.

Basic Qualifications

• 15 or more years of professional experience across some combination of pharma/life sciences, AI/ML product strategy, and product management or strategy roles in technology-led organizations.
• Track record of taking at least one technology or AI initiative from problem framing through executive funding and into delivery.
• Direct involvement in running or building systems for technology or R&D portfolio management, including at least one of: intake, prioritization, funding stages, resource allocation, benefits tracking, or portfolio reviews.
• Hands-on work with modern AI/ML concepts — including LLMs, agentic systems, and digital workers, the ML lifecycle, and data and evaluation patterns — demonstrated through prototyping.
• Prior responsibility for operating as an individual contributor without direct reports and influencing engineering and business teams through frameworks and documented decision criteria.
• Background in product management or product strategy, including opportunity sizing, roadmap development, prototyping, and stakeholder alignment.
• Track record of designing and shipping digital workers and agentic systems that do real operational work, including orchestration, tool use, evaluation, and keeping agents reliable and observable in production.
• Prior responsibility for designing shared libraries of agent skills, prompts, templates, and connectors that multiple teams contribute to and reuse, including versioning, quality standards, ownership, and discovery.
• Direct involvement in operating model design, defining how a portfolio is run day to day — decision rights, gates, cadences, roles, and how work moves between people and digital workers.
• Working knowledge of instrumenting delivery and value data from engineering tools, financial systems, usage telemetry, and AI platforms, and the data modeling and governance needed to turn those signals into continuous, trusted portfolio insight.

Preferred Qualifications

• Prior experience in forward deployed engineering, applied AI, or AI lab/COE role at a pharma, healthcare, or comparable regulated enterprise.
• Background in management consulting, value engineering, or corporate strategy with hands-on technology experience.
• Experience navigating GxP, regulatory, and data-governance requirements in pharma AI deployments.
• Experience codifying product management and user research practices into AI-native tools — such as skills, plugins, and markdown files — and using the command line to automate daily work.