This dashboard is the presentation layer of a data engineering project that spans six interconnected systems: synthetic data generation, a Dataform-managed warehouse in BigQuery, an orchestrated ELT pipeline on Cloud Run, a Pub/Sub streaming intake, a GLM pricing model, and this FastAPI application deployed on Cloud Run.
Every number below comes from a live SQL aggregation against the
dev_claims_analytics
and dev_claims_reports datasets
in BigQuery. The portfolio covers five lines of business across all 32 states of Mexico.
608
Total Claims
$26,944,733
Total Paid (MXN)
$44,317
Avg Severity (MXN)
291
Policyholders
Analysis Views
Vistas de Análisis
Loss Development Triangle
How claim costs develop over time. Shows cumulative paid losses and IBNR reserve patterns by accident year.
Portfolio Health
Underwriting performance trends: claim frequency, pure premium, and loss ratios across coverage lines and years.
Pricing Adequacy
GLM model output comparing predicted required premium vs. actual charged premium to identify mispriced policies.
Geographic Risk Concentration
State-level claim volume and severity. Highlights where portfolio risk is most concentrated across Mexico.
Underlying Data Pipeline
Pipeline de Datos Subyacente
P01
Claims Warehouse
Dataform + BigQuery
P02
Orchestrated ELT
Cloud Run + Scheduler
P03
Streaming Intake
Pub/Sub + Cloud Run
P04
Data Quality
Dataform Assertions
P05
Cost Governance
BQ Slots + Budgets
P06
Pricing ML
GLM → model_scoring
This dashboard reads from the tables that P01 through P06 produce. Code, decision docs, and deployment configs live in github.com/GonorAndres/data-engineer-path.