Quality Control & Process Digitalization

Quality-control data lived in fragmented, often offline processes. Enterprise clients needed a single dashboard for QC operations, plus APIs that could serve large record sets without multi-second waits — and submissions that would not be lost when the network dropped.

Field submissions happen with unreliable connectivity. Reconnect storms — hundreds of pending requests at once — were not an acceptable load pattern.

Designed the architecture and built a Next.js QC dashboard backed by Nest.js REST APIs and PostgreSQL. Redux holds filters, pagination, selected records, and API-driven data. Reusable UI components cover QC workflows end to end. Large-data queries were optimized with indexing, efficient fetching, and Redis caching. An offline-resilient queue stores submissions with no network, then syncs 100–500 pending requests in batches of 10–20 when connectivity returns — so reconnects do not spike the server.

Fetching 10,000 records took ~15 seconds until query plans, indexes, fetch strategy, and Redis caching were in place. Offline submissions needed batching so restored connectivity would not flood the API.

Time to fetch 10,000 records reduced from ~15 seconds to ~5 seconds through query optimization, indexing, efficient data fetching, and Redis caching.

Offline support is incomplete if you only persist locally — you also have to control how those requests come back, or you trade one outage for a self-inflicted spike.