AidBridge: Real-Time Disaster Assistance System (Redis Challenge Submission)

Tenzin Lama — Sat, 09 Aug 2025 18:15:18 +0000

Hi everyone,

This is my submission for the Dev.to Redis Challenge. I've built AidBridge, a real-time emergency response system designed to streamline reporting and managing disaster-related emergencies. The core idea was to showcase the power and versatility of Redis as the central data engine.

GitHub Repo: https://github.com/tenzinl4ma/aidbridge

What I Built

AidBridge is a web application with two main parts:

1. Emergency Reporting Interface

A simple form where people affected by a disaster can submit:

A short message describing the emergency
Their location
Optional contact information

2. Emergency Response Dashboard

A dashboard for responders featuring:

Real-time Feed — Live incoming reports
Full-Text Search — Query emergencies by keyword
AI Pattern Analysis — Automatically matches similar past emergencies
Cache Performance Monitoring — Displays Redis-based AI cache stats

Demo: https://youtu.be/QSXJDcUzshc

How I Used Redis

Redis is the entire backend. It handles real-time streams, full-text search, structured storage, semantic vector matching, AI caching, and usage analytics — all without needing any external systems.

Redis Streams (`emergencies`)

Why: Fast ingestion and real-time display of new reports

How: New reports are added to the stream; dashboard reads with XREVRANGE

Benefit: Ideal for displaying a time-ordered event feed

Redis JSON (`emergency:<request_id>`)

Why: Structured storage for emergency data and AI analysis

How: Complete emergency objects stored as JSON

Benefit: Supports nested structured data, queryable via RedisJSON

RediSearch (`idx:emergencies`)

Why: Full-text and field-specific search

How: Index on Redis JSON keys; search via FT.SEARCH

Benefit: Enables rapid querying of emergency content

Redis Hashes (`emergency:<request_id>:vector`)

Why: Store vector embeddings for semantic similarity

How: SentenceTransformer encodes text to a vector, saved as Redis Hash

Benefit: Enables contextual similarity comparison

Redis Semantic Cache (`ai_cache:<hash>`)

Why: Reduce repeated AI API usage

How: Hash of input checked before calling AI; result stored with TTL

Benefit: Improves speed and efficiency, reduces cost

Redis Stats Counters

Why: Monitor AI cache effectiveness

How: Keys like ai_cache_hits and ai_cache_misses incremented with INCR

Benefit: Enables real-time tracking of cache usage stats

Redis Modules and Features Summary

Feature	Purpose
Redis Streams	Real-time ingestion and feed display
Redis JSON	Structured emergency + AI data storage
RediSearch	Fast full-text and keyword search
Redis Hashes	Store and retrieve vector embeddings
Keys with TTL	Cache AI outputs with expiration
Redis Counters	Track AI cache hit/miss performance

Summary

AidBridge is a fully Redis-powered application. Redis handles everything—real-time data flow, semantic search, structured storage, caching, and monitoring—without needing any external database. This project demonstrates that Redis can serve as a complete, production-grade backend for modern apps.

Thanks for reading.

Repository: https://github.com/tenzinl4ma/aidbridge

Forem: Tenzin Lama