# Momar SOZO - **Event:** [Meta Global AI Developer Hackathon](https://cerebralvalley.ai/e/meta-ai-developer-hackathon) - **When:** Oct 26, 12:00 PM – Nov 4, 9:00 PM (EST) - **Where:** Online - **Team:** [Durga Prashad](https://cerebralvalley.ai/u/Durga_prashad) - **GitHub:** https://github.com/DurgaPrashad/sozo - **Demo video:** https://youtu.be/NEsPdoKwzB8?si=sZD5NrwrWcZo5k9- - **Gallery:** https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery - **Page:** https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/668 SOZO + MOMAR: AI-Powered Scientific Intelligence for Planetary Exploration SOZO (Scientific Observation Zone Operator) is an AI scientific intelligence system designed to help researchers investigate planetary environments, understand scientific observations, discover relevant research and accelerate exploration beyond Earth. Integrated with MOMAR, a planetary intelligence and mission-planning platform, SOZO connects scientific research, multimodal AI, planetary datasets and exploration analysis within a unified ecosystem. Our goal is to make scientific discovery and planetary exploration more intelligent by combining AI-driven research, real planetary datasets, multimodal image understanding, terrain analysis and mission-planning intelligence. Today, planetary scientists and exploration teams work with vast amounts of disconnected satellite imagery, elevation models, geological information and scientific publications. Understanding these datasets, identifying scientifically interesting regions and translating observations into practical exploration plans requires significant manual effort. We are building SOZO + MOMAR to bring these capabilities together. SOZO: AI Scientific Intelligence SOZO acts as the central AI research and scientific intelligence layer. It enables researchers to ask scientific questions, discover relevant publications through Crossref and Europe PMC, explore NASA's Image and Video Library, organize research projects and notebooks, and generate AI-assisted scientific summaries with citations. SOZO can also interpret mission-planning information shared by MOMAR, explain terrain and route metrics, compare exploration alternatives and communicate scientific findings through natural-language interactions. Our vision is to evolve SOZO into an intelligent scientific companion that connects research knowledge, planetary observations and mission intelligence, helping humans investigate complex scientific questions more efficiently. MOMAR: Planetary Intelligence and Mission Planning MOMAR is our planetary intelligence and mission-planning platform for Moon and Mars exploration. It integrates datasets from NASA, ESA and USGS into interactive planetary environments. The platform supports 3D planetary visualization, geological exploration, elevation and slope analysis, terrain-risk assessment and the comparison of potential exploration routes. Its existing workspace includes detailed Mars environments covering Jezero and Gale craters, alongside a lunar south-pole exploration environment. Together, MOMAR and SOZO connect planetary data with AI-assisted scientific investigation, allowing researchers to explore planetary regions, understand available evidence and investigate potential exploration opportunities. What We Plan to Build with Meta AI During the hackathon, we plan to extend SOZO using Meta's available models through the Meta Model API. Muse models: Explore their capabilities for scientific reasoning, research assistance and multi-step interpretation of planetary exploration context. SAM 3.1: Develop image-segmentation workflows to identify and isolate scientifically relevant regions in planetary imagery, subject to compatibility with the selected datasets. Muse Image: Explore AI-assisted visual explanations and research illustrations for planetary observations. Multimodal scientific intelligence: Connect planetary images, terrain context and scientific literature within a unified SOZO research workflow. Our target experience is straightforward: a researcher selects a region on the Moon or Mars, investigates its visual and terrain characteristics, asks SOZO scientific questions, retrieves supporting research and compares exploration possibilities within one environment. We aim to combine AI-generated interpretation with traceable scientific sources and deterministic terrain calculations. AI-generated explanations will not be presented as verified scientific measurements, and mission-planning decisions will remain subject to human review. Technical Foundation and Execution We have already developed a working SOZO research platform and the MOMAR planetary exploration environment. The existing technical foundation includes scientific literature discovery, NASA image archive search, research projects and notebooks, cited AI synthesis, deterministic numerical analysis, REST APIs, 3D planetary visualization, planetary dataset integration, terrain analysis and an AI companion connected to MOMAR. The hackathon will focus on integrating Meta's models into SOZO and demonstrating an end-to-end multimodal planetary research workflow. Our Vision We believe the next generation of scientific discovery will be powered by intelligent systems that can connect scientific knowledge with observations of the physical world. SOZO provides the scientific intelligence. MOMAR provides the planetary intelligence and exploration environment. Together, we aim to build an intelligent ecosystem that accelerates scientific discovery, advances planetary research and helps prepare humanity for exploration beyond Earth. Building intelligence for humanity's multiplanetary future. ## More from Meta Global AI Developer Hackathon - [FounderFlo Ops](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/665) - [Phoenixon](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/666) - [Djinn Council](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/667) - [Cortex AI](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/669) - [Ontology-Physiology Lab](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/670) - [Maina Lewis](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/671) --- Markdown version of https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/668. Site index for agents: https://cerebralvalley.ai/llms.txt · full text: https://cerebralvalley.ai/llms-full.txt