Astro is a cloud-native RAG (Retrieval-Augmented Generation) platform built to turn dense NASA bioscience PDFs into a searchable knowledge base you can actually talk to. It uses serverless backend tech to ingest, chunk and semantically index research documents, then delivers AI-grounded answers through a space-themed chat interface — with source citations and optional web search for extra context. Born at the NASA Space Apps 2025 NYC Hackathon, Astro won an award for its full-stack use of vector search, Cloudflare Workers, and generative AI to make complex research accessible without slogging through mountains of text. ([GitHub][1])
Built With
- cloudflare
- gemini
- javascript
- langchain
- postgresql
- rag
- react
Log in or sign up for Devpost to join the conversation.