PAVITHRA NANNA

Built React and TypeScript feature modules for an internal platform and integrated them with Node.js services via versioned REST API contracts. Designed a Redis caching layer for high-traffic read endpoints, reducing median response times from 320ms to 90ms during peak load. Improved user-facing reliability by tracing recurring client-side failures to silent API errors and rebuilding React error handling to show actionable messages instead of blank states. Automated Azure deployments with GitHub Actions pipelines, adding staged rollouts and rollback steps to cut release time by 45%. Instrumented Azure services with structured logging and alerts, then led incident triage for production degradation impacting the platform’s write path.

Built React and TypeScript screens for a retail e-commerce checkout flow, integrating them with Express endpoints for cart and order state. Integrated third-party payment and shipping APIs into a healthcare payer portal using a reusable client wrapper to replace brittle point-to-point calls. Refactored a Node.js reporting service used by an insurance claims intake team by splitting a monolithic handler into tested, composable modules. Selected MongoDB document modeling for a travel logistics catalog after evaluating that varied product attributes made relational joins costly. Added Jenkins build stages for a media client web app, raising automated test coverage to 85% before releases.

Integrated legacy backend services with a modern React front end for a banking client’s account servicing portal using REST API adapters. Migrated SQL Server reporting queries behind a financial services client dashboard by rewriting costly joins and adding indexes to reduce runtime. When the existing Java service could not expose needed fields, designed an adapter layer rather than forking the service, keeping teams on one codebase. Documented integration contracts and ran sprint demos for an energy utilities client to keep backend and frontend teams aligned on delivery.