This job is about bringing AI ideas to life right at the centre of Pfizer. The Office of the Chief Executive Officer (OCEO) shapes the company's strategy, how it talks to the world and where it invests in global health, and it runs the CEO's own office. The ideas you take forward end up in that work. It might be the briefing the CEO reads before meeting another company's chief executive, the signals behind a strategic call, or a message tested before Pfizer says it in public. It's rare for product work to sit this close to the decisions.
Ideas come in all shapes. Often it's just something a colleague has thought of. Sometimes it's a prototype they built with an AI assistant, or a pilot an outside partner ran. You decide which ones are worth doing, and you take the good ones all the way to something properly supported, governed and used by the people it was for.
The team is small and senior, five people in all. Your closest partner is the Director, Engineering Lead. You own what gets built and why, they own how it's built, and most of the big calls you'll make together.
You'll work with the leaders of OCEO's four groups: Strategy and Consulting, Corporate Communications, Global Health Investments and Executive Operations. You get a real say over what the team builds and what it leaves alone, and you'll need the evidence and the standing to say no. We're here to help OCEO build for itself rather than to take orders. Success looks like an Office that builds on shared capabilities, where whoever had the idea stays with it until it ships.
Role Responsibilities
Meet the person with the idea early In the first few days, sit down with whoever brought the idea in and turn it into a short brief they sign off. It says what should change, for whom, and how we'll know it worked, so we start from an outcome and not a feature list. The originator stays with the idea until release, and it's your job to make sure that actually happens.
Decide what's worth doing Ideas turn up as a few lines in a chat, a prototype or a pilot. You weigh what each would be worth, how much work it would take to make it production-grade, and whether anyone would really use it. Back the ones that earn it. Reshape the ones that are nearly there. Stop the rest early, and be open about why, so the team's time goes where it counts.
Own the ROI definition For every use case, you define what return on investment (ROI) means for the AI agents behind it: what they cost to build and run, what they should give back, and how we'll know. Get it agreed with the originator before work starts. The team uses it to decide what to back, and OCEO uses it to judge whether the work paid off.
Look for the cheapest way Before anyone designs anything, check what already exists in OCEO and on Pfizer's enterprise platforms. A lot of ideas are best served by reusing or extending something we already have, and a new build has to justify itself. The hard part is holding that line when a senior stakeholder would rather have a tool of their own.
Own the spec You own the specification and review stages of how the team builds. Write briefs and acceptance criteria tight enough for an AI-assisted build to work from. When the work comes back, judge it against the outcome in the brief, not the wording of the ticket. Loose specs make for loose products.
Work with the groups You'll work with the leads of the four groups and their teams, and with the colleagues who own each solution once it's live. Keep expectations honest about what's actually coming, agree how originators get credit for the time they put in, and raise friction early instead of at release.
Make sure people use it Plan adoption with each group, find local champions, and help colleagues do more for themselves next time. That way OCEO gets a bit more capable with every idea, and the know-how doesn't all sit with the team.
Measure what changes Once work starts, track each use case against its ROI definition and report regularly: which ideas…