Senior Financial Analyst, FP&A (AI First)

Department: Operations

Location: Remote, Canada or US

Position Type: Full-Time, Permanent

Hours: Monday-Friday, 9AM-5PM (in your time zone)

About the Role

Unbounce is looking for a Senior Financial Analyst, FP&A (AI First) to join our FP&A team. You'll run our budgeting and forecasting cycles, prepare monthly reporting, support the accounting team through month-end close, and partner with teams across the business to answer financial questions.

This is an AI-native role: the outputs are familiar (an annual budget, a rolling forecast, a monthly reporting package, board metrics), but you will spend a meaningful share of your time building the systems that produce them rather than producing them by hand. Every employee at Unbounce has Claude, and our billing and accounting stack was rebuilt this year on Recurly and Rillet, so the data layer underneath you is new rather than inherited.

Concretely, when a recurring analysis takes four hours by hand, we expect your instinct to be to build the thing that does it in four minutes next month, and then to be the person who can tell whether its answer is right. The second half matters more than the first. Automation without judgment is worse than no automation, and this role owns both.

We're after someone dependable and detail-oriented who communicates clearly and whose first instinct on a repetitive task is to automate it. You don't need to have done every item on this list. If you're strong in most areas and willing to learn the rest, we'd like to hear from you.

What You’ll Be Doing

Build and operate AI-driven finance workflows

• Design, build and maintain the AI agents and automated workflows that carry the recurring FP&A load: variance analysis, first-pass commentary, cohort and retention analysis, board metric refreshes, and the routine data pulls underneath all of them.
• Use AI as a default part of your workflow, across analysis, modeling, reporting, and commentary, within our team practices and data-handling guidelines.
• Use Claude, which every employee here already has, to speed up analysis, automate recurring reporting, draft first-pass variance commentary, and pressure-test your own assumptions before numbers go out.
• Build the context these agents need in order to be reliable: metric definitions, documented data lineage from source system to report, and explicit rules for how each number is constructed. An agent is only as good as the context it can reach.
• Decide what should and should not be automated. Draw and defend the line between work an agent runs unsupervised, work an agent drafts for human review, and work that stays fully manual.