HelpHive

Inspiration

We started with a simple observation: aging in place is getting harder. Everyday tasks like carrying groceries, climbing stairs, or navigating technology can quickly become safety risks for elderly neighbors. Meanwhile, communities are full of people willing to help, but there just isn't an easy, trusted way to connect them. HelpHive was born from that gap. Our mission is to turn neighborhoods into built-in support networks using AI + human kindness.

What it does

HelpHive is a lightweight community assistance platform that connects elderly and differently-abled neighbors with verified local volunteers. Users submit requests through voice, text, or photos. Claude instantly transforms that messy input into a structured task card:

• A clean 2–5 word title
• A clear, rewritten description
• Smart classification: category, urgency, people needed
• Extracted task types
• AI-powered prioritization for volunteers Volunteers nearby receive real-time alerts, view tasks on an interactive map, and can safely accept and complete them. HelpHive makes getting (or giving) help feel as natural as sending a message.

How we built it

• React + Firebase for fast, reliable real-time task updates
• Claude 3 Sonnet for request parsing, triage, and prioritization
• Google Maps API for location-aware volunteer matching
• Client-side caching + quotas to control API usage
• Firestore for secure, real-time task storage
• Accessibility-first UI designed for elderly users We built the full end-to-end experience in a single day: request → AI parsing → prioritization → volunteer routing → safe completion.

Challenges we ran into

• Getting Claude to consistently produce strict JSON with short titles
• Balancing natural-language input with the rigid structure needed for tasks
• Configuring Firebase + Google Maps keys cleanly inside a React environment
• Creating a safe, trustworthy experience without adding user friction
• Handling geolocation, map rendering, and volunteer proximity logic under time pressure

Every challenge helped us refine the core value: simplicity + human-centered design.

Accomplishments that we're proud of

• A functional, end-to-end platform delivered in under 10 hours
• An AI parser that turns unstructured input into actionable, high-quality tasks
• Real-time volunteer matching with safety checks
• A polished, intuitive interface built for older users
• A system that feels fast, simple, and genuinely impactful

We didn't just build features — we built a product that could exist tomorrow.

What we learned

• How to design AI assistance that feels trustworthy and human
• Why prompt engineering must be brutally specific for structured outputs
• How to build real-time, map-based workflows with Firebase
• The importance of safety, clarity, and minimalism in community tools
• How to scope smartly under intense hackathon pressure

What's next for HelpHive

HelpHive has a clear path to becoming a real community infrastructure layer.

Next steps include:

• Mobile app development for iOS and Android to increase accessibility and enable on-the-go assistance
• Direct integration with local emergency services including police, fire departments, and emergency medical services
• Partnerships with NGOs, hospitals, and community organizations to expand volunteer networks and build trust
• Senior vulnerability scoring to improve prioritization and identify high-risk individuals
• Reputation and verification system for volunteers with background checks and community ratings
• Enhanced safety monitoring including fall detection, audio check-ins, and real-time status updates
• Multi-language support to serve diverse communities
• Backend proxy architecture for secure, scalable AI usage and data protection

Our long-term vision: HelpHive becomes the community's front door for assistance, transforming neighborhoods into responsive, resilient ecosystems of support.

Built With

• anthropic-claude-api
• cloud-firestore
• create-react-app
• firebase-authentication
• firebase-storage
• google-maps-javascript-api
• google-oauth-2.0
• google-places
• googlemaps/js-api-loader
• javascript-(es6/jsx)
• jest-&-react-testing-library
• react-18
• react-router-dom