Exploring how life adapts beyond Earth with conversational research, graph intelligence, and cinematic UI moments.
BioQuery was built for the NASA Space Apps Challenge, turning more than 600 bioscience publications into an AI research engine for mission planners and scientists.
We wanted natural-language discovery, so Astro-Voyagers fused RAG, knowledge graphs, and Cohere embeddings to help teams ask ambitious questions about life beyond Earth.
Context panel
Each journal entry is linked to missions, organisms, and outcomes so insights travel faster.
Alternating layouts keep the scroll cinematic while the videos showcase live product moments.
BioQuery listens to researcher intent and runs Retrieval-Augmented Generation so scientists can ask conversational questions about biology in space missions.
Grounded answers blend live search with NASA's bioscience corpus.
A custom embedding pipeline pairs Cohere vectors with graph edges so you can orbit through related experiments, missions, and organisms in seconds.
Graph hops reveal latent links between microgravity studies.
The interface ships with streaming summaries, timeline pivots, and exportable research briefs that keep multidisciplinary teams aligned.
Built with agentic helpers so analysts never leave the page.
Recognized for pushing human exploration with grounded research tooling.
"BioQuery started as a weekend sprint but turned into a vision for how AI can accelerate biology for deep space."
Awards panel commentary
Publications processed
Core AI pipelines
Galactic Impact Award
BioQuery let judges navigate complex datasets without touching SQL, shining a path for future mission tooling.
Cohere Embeddings - RAG - Knowledge Graphs
Node.js - Vercel Functions
Next.js 15 - TailwindCSS - Framer Motion
Vercel
Hackathons test speed, but BioQuery reminded me to design for lasting curiosity.
"BioQuery is proof that passion, clarity, and AI copilots can compress months of research tooling into a single weekend."
David Zahemen Yeboah - Builder