Inspiration Animal rescue and adoption support is fragmented across clinics, shelters, and volunteer networks. In emergencies, people lose critical time searching across multiple sources. We wanted one assistant that gives immediate, actionable help.
What it does AI Pet Rescue and Adoption Co Ordinator provides:
Emergency support: finds nearest emergency vet options and rescue contacts by location. Adoption matching: suggests suitable pets based on lifestyle preferences. Resource discovery: surfaces shelters, urgent needs, and support options. Conversational guidance: gives practical next steps in simple chat form. How we built it Built a FastAPI backend with a chat endpoint and task-specific APIs. Indexed rescue data in Elasticsearch with geo-enabled mappings. Added AWS Bedrock integration for dynamic intent routing and response generation. Implemented an interactive web UI with quick prompts, map view, and result cards. Added scripts for data conversion, seeding, geo-query validation, integration testing, and demo evidence generation. Challenges we ran into Model/version and region differences in Bedrock (older model IDs retired). Elasticsearch serverless differences from classic clusters. Balancing deterministic reliability with dynamic AI routing. UI usability issues (chat overflow/scroll behavior and layout comfort). Location handling beyond Bengaluru while dataset remained city-focused. Accomplishments that we're proud of Delivered a working end-to-end application, not just a prototype API. Combined live Elasticsearch search with Bedrock-powered dynamic chat flow. Built emergency, adoption, and resource journeys with actionable outputs. Added demo-ready assets: script, evidence captures, backup plans, and runbooks. Improved UX with editable quick queries, one-click demo mode, and map integration. What we learned Real-world rescue tools need both AI flexibility and strict fallback safety. Data quality and coverage matter as much as model quality. Geospatial indexing is powerful for emergency response use cases. Bedrock and Elastic integration can move from idea to demo very quickly with the right architecture. Good demo engineering (scripts, evidence, fallbacks) is as important as core features. What's next for AI Pet Rescue and Adoption Co Ordinator Expand data coverage from Bengaluru to multiple Indian cities/states. Add automated ingestion pipelines for shelters and veterinary directories. Improve semantic matching with vector search and richer preference profiles. Add multilingual support and WhatsApp/mobile channels. Add observability dashboards, reliability metrics, and production hardening.
Built With
- built-with:-language:-python
- conversion
- css
- data
- fastapi
- geo-enabled-index)-cloud-services:-aws-+-elastic-cloud-geocoding/maps:-openstreetmap-nominatim-(geocoding)
- google-maps-embed-(ui-map-view)-validation/testing:-pytest
- html
- javascript-backend-framework:-fastapi-asgi-server:-uvicorn-ai/llm-platform:-aws-bedrock-(anthropic-claude)-search-&-database:-elasticsearch-(elastic-cloud
- json/ndjson
- testclient
- tooling:
Log in or sign up for Devpost to join the conversation.