Salt Chamber Data Collection and AI Generated Reporting

Developed an internal web dashboard that consolidates salt-spray corrosion testing and pretreatment line readings into a single real-time view for the QA team. The project involved designing the data model, API layer, and front-end interface, enabling automatic batch grouping of weekly chamber logs, overdue-test detection, and spec-aware visualisation of daily pretreat chemistry. The system integrates directly with existing N8N mobile forms, operators continue submitting weekly chamber logs and daily pretreat readings as before, with submissions flowing straight into a PostgreSQL database that the dashboard reads through tRPC. Key features include a custom batch-grouping engine (14-day gap windowing, 60-day incomplete threshold), material-aware spec tolerance charts for Aluminium and Zinc pretreat lines, and a same-origin Drive photo proxy that allows network-restricted factory tablets to view inspection images. The application is deployed as a multi-stage Docker container behind an nginx reverse proxy with TLS termination, running alongside the existing Postgres instance. This solution improves visibility, traceability, and out-of-spec response time for QA across APLM production environment, and the same framework is now being replicated for other data-capture processes across the business.