Database & Analytics Modernization for Sunny

Challenge.

Sunny Benefits operates at the intersection of fintech and benefits administration, a space where data correctness and timeliness directly drive compliance posture, broker confidence, and end-customer trust. The team needed a single data foundation capable of serving three distinct audiences from the same source of truth: internal product analytics, benefits operations workflows, and customer-facing insights embedded in their application. Standing up bespoke pipelines per use case had become untenable — each new dashboard or report introduced drift between systems, slowed engineering velocity, and made auditability harder to defend during SOC and partner reviews.

On the runtime side, deployment patterns had diverged across services as the product matured. Some workloads ran on legacy compute, others on early container experiments, and there was no consistent way to scale with the sharp, predictable usage spikes tied to open enrollment windows, payroll cycles, and end-of-quarter reconciliation. Engineering spent disproportionate time on capacity firefighting rather than feature work, and deploys carried more risk than the team was comfortable with given the regulated nature of the data flowing through the platform.

Solution.

BUILDSTR designed a containerized application platform on AWS paired with a partitioned analytics layer tuned for both operational dashboards and ad hoc finance and benefits queries. The container platform was built around predictable auto-scaling policies aligned to enrollment-driven traffic patterns, with blue/green deployment paths, automated health gating, and standardized service templates so new workloads inherit the same guardrails by default. Workload isolation, secrets management, and network segmentation were structured to fit fintech-adjacent expectations on least-privilege access and change traceability.

The analytics layer separates hot, customer-facing queries from heavier internal analysis through partitioning and workload routing, so a long-running finance query can't degrade the experience of a broker pulling a live dashboard. Data contracts and a governed semantic layer ensure that metrics like enrollment counts, premium totals, and contribution splits reconcile across product, finance, and customer-facing surfaces. CI/CD, observability, and access controls were standardized end-to-end — every change is reviewable, every query is attributable, and every deploy is reversible — fitting the auditability and change-management expectations of a fintech-adjacent ISV.

Results.

• 3x query speedup vs. the prior reporting stack, with consistent performance through enrollment peaks and month-end close.

• 45% reduction in deployment lead time on the containerized platform, with safer rollback paths and fewer change-related incidents.

• 5+ benefits insights dashboards delivered to end customers from a single governed data layer, eliminating per-customer pipeline sprawl.

• 25% lower infrastructure cost at steady state through right-sized auto-scaling and workload isolation between operational and analytical traffic.

• Audit-ready change history across application and data layers, accelerating partner and compliance reviews.