Offline-First Engineering for Disaster-Ready Applications

Most applications assume the internet is always available. In disaster response, remote operations, and field deployments, that assumption fails quickly.

This talk presents the real-world engineering journey behind building an offline-first crisis coordination platform designed for disaster-prone regions of Nepal. What began as a simple requirement—”make it work offline”—evolved into a distributed systems challenge involving operation logs, synchronization protocols, conflict resolution, peer-to-peer replication, logical clocks, retry mechanisms, and observability under unreliable connectivity.

Through architecture diagrams, implementation lessons, production failures, and field-tested examples, attendees will see why offline-first is not merely a feature but a fundamentally different approach to system design. The session explores how traditional client-server assumptions break down when networks become intermittent, devices become authorities over local observations, and multiple users modify data concurrently.

Attendees will learn practical patterns for designing resilient systems that continue operating during outages, synchronize safely when connectivity returns, and preserve user trust under failure. While inspired by disaster-response software, the principles discussed apply equally to mobile applications, collaborative tools, edge computing platforms, IoT systems, and any application operating in unreliable network conditions.

If your users cannot afford to stop working when the internet disappears, this talk is for you.