SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Salary: USD 156,000 - 195,000 / annual
Airbnb's Fraud & Safety Delivery (FSD) organization is building AI-native operational infrastructure to keep the platform safe for millions of hosts and guests. This role sits within Scaled Services, the foundational team that owns how FSD launches, measures quality, builds tools, and uses data.
As Senior Program Manager, Quality Assurance & Standards, you will build FSD's quality program from the ground up—not an audit checklist function, but a comprehensive measurement system that tells the organization how accurate its decisions are, where quality breaks down, and what to change. You'll design and stand up this framework while running the quality reviews the organization depends on today, reporting to the Senior Manager, Scaled Services.
You will own three core outcomes:
1. A defensible quality framework: One documented model measuring both workflow adherence and decision quality, with published standards, thresholds, and severity definitions across every risk domain.
2. Objective, centralized governance: Serve as the impartial quality authority across FSD, setting consistent standards that hold up across domains, sites, and geographies.
3. Measurable improvement in decision accuracy: Fewer incorrect enforcement outcomes, diagnosed at root cause and fixed at the source, without adding cost or handling time.
Key responsibilities include:
- Own the quality program end-to-end, including delivery through partner sites, and set the strategy and operating rhythm for how quality delivers against FSD's roadmap.
- Design a two-layer quality model measuring both workflow adherence and enforcement decision accuracy; establish how to determine correct outcomes for judgment-based decisions through expert review panels, blind double-review, and benchmark case libraries.
- Build a severity-weighted error taxonomy and risk-weighted sampling methodology; author quality guidelines and publish performance bars, thresholds, and escalation paths.
- Run a calibration program across internal teams and partner sites, keeping standards current as policy, product, and workflow evolve.
- Deliver quality reviews across FSD on consistent methodology and cadence; own User Acceptance Testing (UAT) for new launches.
- Diagnose root causes behind reversed decisions and produce diagnostic quality reporting that leadership and cross-functional partners can act on.
- Extend quality measurement to AI and model-driven decisions; define accuracy standards required before automation; partner with Product, Engineering, and Data Science to feed labeled quality data back into detection models.
- Trace quality findings to their source and drive fixes with owning teams; improve decision accuracy without adding handling time or review layers.
Fraud and Safety Delivery is judgment work at scale. Specialists weigh incomplete evidence and make consequential calls about accounts, listings, and bookings thousands of times daily across many risk domains globally. Precision matters in both directions: a missed signal puts the community at risk, and incorrect enforcement affects someone's livelihood or travel plans.
REQUIREMENTS:
- 8+ years of experience in quality assurance, operations, risk, audit, or program management within a high-volume, high-stakes decisioning environment.
- Proven experience driving quality delivery through partner or vendor teams you don't directly manage.
- Proven experience building a quality program from scratch (not just running one): you've designed sampling methodology, scorecards, error taxonomies, calibration processes, and governance from a blank page.
- Demonstrated experience measuring quality in subjective, judgment-based work; adjacent fields welcome: fraud and financial crime, content moderation, claims review, credit decisioning, regulatory compliance, or trust and safety.
- Strong grasp of measurement rigor: understanding of sampling design, statistical significance, and inter-rater reliability; ability to defend a quality number to a skeptical executive or partner site.
- Strong data literacy: ability to pull and interrogate your own data, work fluently with analytics partners, and build diagnostic reporting.
- Exceptional writing skills: guidelines must be clear enough that reviewers across regions and partner sites reach the same conclusion.
- Strong cross-functional influence skills: ability to tell teams their workflows, policies, or models are producing wrong outcomes without losing them as partners.
- Experience evaluating model or AI-assisted decision quality is a strong plus.
- Experience in Trust, Fraud, or Safety is a plus, but not required; the right instincts for precision, user impact, and intellectual honesty matter more.
