Business Intelligence & Data Visualization (Power BI/Tableau/Cognos)
This area assesses how you transform requirements into usable, performant, and secure BI products. Expect to discuss data modeling for BI, DAX/M or calculated fields, row-level security, and performance tuning strategies. You should be able to translate a BRD into a semantic model, define KPIs unambiguously, and produce an executive-ready dashboard narrative.
- Be ready to go over:
- Data modeling for BI: Star vs. snowflake schemas, semantic layer design, grain and conformance
- DAX/M (or calculated fields): KPI logic, time intelligence, incremental refresh strategies
- Security & deployment: RLS/OLS, workspace strategy, deployment pipelines, usage monitoring
- Advanced concepts (less common): Composite models, DirectQuery vs. Import vs. DirectLake (Fabric), Cognos Framework Manager, Fabric Lakehouse
- Example questions or scenarios:
- “Design a Power BI model and dashboard for Commercial Property loss ratio by product, broker, and region. How do you define earned premium and control data access?”
- “Your report is slow against a large Snowflake model. What steps do you take to diagnose and resolve performance?”
- “How would you implement RLS for a broker hierarchy while preserving accurate roll-ups for regional leaders?”
SQL, Data Modeling & Warehousing
We evaluate your depth in SQL, core modeling approaches (3NF, dimensional, Data Vault), and your ability to profile and remediate data quality. Expect window functions, CTEs, joins, and reasoning about SCD Type 2 and lineage. Familiarity with Snowflake features (e.g., clustering, tasks/streams) is valuable.
- Be ready to go over:
- SQL proficiency: Aggregations, window functions, anti-joins, performance considerations
- Modeling tradeoffs: 3NF vs. star schema vs. Data Vault for analytics and governance
- Data quality & lineage: Profiling, reconciliation, incident handling, auditability
- Advanced concepts (less common): Hubs/links/satellites (Data Vault), surrogate keys, ER modeling tools, Snowflake optimization
- Example questions or scenarios:
- “Write a SQL query to calculate 12-month rolling loss ratio by product and geography.”
- “When would you choose Data Vault patterns over dimensional modeling at AIG?”
- “How do you reconcile policy counts that differ between source and warehouse, and how do you document the decision trail?”
Analytics & Insurance Domain
You don’t need to be an actuary, but you should understand the core insurance analytics vocabulary and how metrics drive decisions. We’ll probe your grasp of loss ratio, expense ratio, combined ratio, frequency/severity, and how to build trustworthy KPI definitions end-to-end.
- Be ready to go over:
- Core metrics & definitions: Earned vs. written premium, exposure, IBNR sensitivity, calendar vs. accident period
- Operational analytics: Underwriting funnel, broker/channel performance, claims triage and cycle time
- Decision support: Pricing, reserving insights, portfolio steering, trend analysis
- Advanced concepts (less common): GLM basics, credibility concepts, geospatial exposure, catastrophe enrichments
- Example questions or scenarios:
- “Explain earned vs. written premium to a non-technical stakeholder and how it affects loss ratio.”
- “Your claims severity spikes in one region—how do you investigate and present findings?”
- “Design a star schema to analyze policy, claim, and broker performance together.”
Data Governance, Security & Compliance
AIG operates in a heavily regulated environment. You will be assessed on data classification, PII handling, RLS/OLS, catalog/lineage practices, and how you embed governance in the SDLC. Expect practical questions about change management and audit readiness.
- Be ready to go over:
- Access control & security: RLS/OLS, least privilege, secrets management, audit trails
- Data governance: Definitions stewardship, quality rules, catalog/lineage, issue management
- SDLC controls: Promotion processes, approvals, rollback plans, usage monitoring
- Advanced concepts (less common): GDPR/CCPA applicability, SOX implications for reporting, retention policies
- Example questions or scenarios:
- “How would you design access for regional leaders, brokers, and auditors viewing the same dashboard?”
- “Describe how you’d document KPI definitions and lineage to support an internal audit.”
- “A metric needs to change in production—what process ensures continuity and traceability?”
Communication, Stakeholder Management & Delivery
We measure how you elicit requirements, align on definitions, manage tradeoffs, and ship. You should demonstrate product thinking: clear scoping, prioritization, iteration, enablement, and adoption tracking—ideally using Agile practices and tools like Rally.
- Be ready to go over:
- Requirements to roadmap: Turning BRDs into product backlogs, MVP scoping, acceptance criteria
- Alignment & enablement: Workshops, stakeholder maps, training, documentation
- Adoption & success: Usage analytics, feedback loops, versioning, deprecation plans
- Advanced concepts (less common): Value realization metrics, OKRs for BI products, change management playbooks
- Example questions or scenarios:
- “Walk us through a complex dashboard you led from BRD to deployment. How did you drive adoption?”
- “Two executives disagree on a KPI definition—how do you reconcile and proceed?”
- “How do you structure a 10-minute readout to senior leaders on portfolio performance?”
