Nepal Disaster Relief
Disaster information, live situation monitoring and official-source data automation.
Overview
What I built.
A public disaster-information platform with disaster archives, situation monitoring, missing-person information, funding/transparency views and automated data ingestion from official sources.
Challenge
The problem.
Keep public-facing disaster information useful and responsive when upstream sources are slow, inconsistent or updated on different schedules.
Contribution
What I did.
Developed and stabilised Next.js/FastAPI/PostgreSQL pages and APIs for situation data, missing persons, disaster history and transparency.
Worked on automated official-data workers and update flows for changing public information.
Introduced caching, timeouts and stale-value preservation patterns to reduce failures caused by slow upstream services.
Investigated API latency, pagination, first-load failures, data-shape bugs and production regressions.
Maintained deployment and performance while preserving previously fixed features during iterative updates.
Architecture
System flow.
Highlights
Outcome
A resilience-focused public-information system designed to remain useful even when external data sources are imperfect or slow.