STARSCAN.AI: NASA Space Apps Challenge 2023
Challenge: STAR: Revolutionizing Technical Standards with AI
STARSCAN.AI: NASA Space Apps Challenge 2023
Challenge: STAR: Revolutionizing Technical Standards with AI
About STARSCAN.AI
Our Mission
Our Expertise
Our Mission
It is an AI-powered Aerospace Documentation Explorer. NASA has a vast repository of millions of technical and scientific documents which are complex and multidimensional. Inconsistencies in metadata are prevalent, and missing or irrelevant data pose great challenges especially when we are trying to understand the technical scientific requ
It is an AI-powered Aerospace Documentation Explorer. NASA has a vast repository of millions of technical and scientific documents which are complex and multidimensional. Inconsistencies in metadata are prevalent, and missing or irrelevant data pose great challenges especially when we are trying to understand the technical scientific requirements. Our mission is to develop AI systems to streamline the querying and recommendations of such technical guidance.
Our Team
Our Expertise
Our Mission
Our team is comprised of experienced software developers, AI engineers, designers, and project managers who are dedicated to delivering exceptional results. We work collaboratively with our clients to ensure that their unique needs are met and that they receive the best possible service.
Team Leader: Sreekanth Gopi, USA
AI Engineer: Louis Z
Our team is comprised of experienced software developers, AI engineers, designers, and project managers who are dedicated to delivering exceptional results. We work collaboratively with our clients to ensure that their unique needs are met and that they receive the best possible service.
Team Leader: Sreekanth Gopi, USA
AI Engineer: Louis Zhang, Indonesia
Design AI Engineer: Devananda Sreekanth, USA
Junior AI Engineer: Kunal Arora, India
Our Expertise
Our Expertise
Our Expertise
At STARSCAN.AI, we have expertise in a wide range of software development areas, including AI system development, ML engineering, Data Science, web development, mobile app development, and custom software development. The team behind STARSCAN.AI boasts a robust blend of expertise, ensuring the project's success. Leading the team is Sreeka
At STARSCAN.AI, we have expertise in a wide range of software development areas, including AI system development, ML engineering, Data Science, web development, mobile app development, and custom software development. The team behind STARSCAN.AI boasts a robust blend of expertise, ensuring the project's success. Leading the team is Sreekanth Gopi, who holds an MSCS in AI. His past accolades as a hackathon winner demonstrate his ability to drive innovation and lead projects to fruition. Louis Zhang, the AI Engineer, brings to the table his prowess in AI coding. His past victories in hackathons further attest to his proficiency in the domain. Kunal Arora also is past hackathon winner. Rounding out the team is Design Engineer Devananda Sreekanth, a BSCS holder with a passion for hackathons. Together, their combined skills and experiences position them uniquely to revolutionize technical standards with AI.We use the latest technologies and tools to ensure that our clients receive cutting-edge solutions that drive their business forward.
STARSCAN.AI: Empowering Your Technical Document Processing
The Problem - NASA technical Document
- Complexity & Multidimensionality
- Inconsistent Metadata Use
- MISSING & IRRELEVANT DATA
- Size & Computation Intensity dynamic
THE SOLUTION
- AI Solution to analyze Technical/scientific documents accurately and efficiently.
- Semantic vector embeddings
- Standardized data representation.
- Threshold-based SIMILARITY CHECK
- Efficient chunk-based processing.
- A complex time-consuming challenge dealt intelligently with LLM technology.
THE PIPELINE
- User Input: Streamlit-powered querying and document upload.
- PDF Processing: Load, split, and convert NASA documents.
- Embedding Generation: Convert text to vector embeddings with SentenceTransformer.
- LLM Setup: Clarifai's GPT-4 integration for deep text analysis.
- Data Retrieval: Use Pinecone for efficient data fetching.
- Structured Responses: PromptTemplates guide LLM to generate structured answers.
DATA SECURITY METHODOLOGY
- API Key Encryption: User controls using TOML-encrypted key protection and secret API keys.
- Relevance-Driven Data Filtering: Displays data only if similarity score exceeds eg: 0.5.
- Guided AI Responses: PromptTemplate directs LLM, mitigating unintended data exposure and avoid LLM hallucinations.
- Negative Query Management: Handles off-topic questions, preventing irrelevant outputs.
- Targeted Data Processing: Focuses solely on NASA Technical documents, minimizing data risks.
- Structured Output: Prompt engineering for output in a given specific content format.
AI Platforms applied
- Clarifai: Enables GPT-4 LLM integration.
- BeautifulSoup: Extracts NASA Technical Bulletin links | Fetches and segments PDFs.
- Langchain: PromptTemplate for LLMs | Links document retrieval, vectorization, and LLM interaction.
- HuggingFace & Pinecone: Transforms text to vectors and stores in Pinecone vector database for quick retrieval.
- GPT-4: Analyses scientific text, fills missing data, removes irrelevant data, formulate reports.
- Streamlit: Offers interactive UI for querying NASA documents on a web app.
Data - NASA documents
- NASA Standards
- NASA Standard 5018, Strength Design and Verification Criteria for Glass, Ceramics, and Windows in Human Space Flight Applications
- Sample NASA standards document that your new tool could analyze
- NASA Engineering and Safety Center (NESC) Technical Bulletins
- A technical bulletin that captures critical knowledge in the form of new engineering information or best practices in a one-page format.
- NESC Technical Bulletins Summary
- NASA Lessons Learned
- The NASA Lessons Learned system provides access to official, reviewed lessons learned from NASA programs and projects. These lessons have been made available to the public by the NASA Office of the Chief Engineer and the NASA Engineering Network. Each lesson describes the original driving event and provides recommendations that feed into NASA's continual improvement efforts via training, best practices, policies, and procedures
- NASA Technical Reports Server (NTRS)
- NTRS provides access to aerospace-related citations, full-text online documents, and images and videos. The types of information include technical reports, conference and meeting papers, journal articles, patents, and contractor and grant reports. A public interface contains scientific and technical information created or funded by NASA, and its predecessor the National Advisory Committee for Aeronautics (NACA). The full NASA scientific and technical information (STI) collection, including documents collected from U.S. and international sources, is available to qualified users via the NTRS-Registered delivery platform. That platform includes over four million bibliographic records and a growing number of full-text documents.
- NASA Knowledge Management
- The Office of the Chief Knowledge Officer disseminates NASA lessons learned in a variety of formats, including NASA knowledge-sharing publications, stories, and events. This site also includes points of contact for NASA's various knowledge-producing activities.
- NASA Scientific and Technical Information (STI) Program
- The mission of the NASA Scientific and Technical Information (STI) Program is to acquire, disseminate, and archive NASA STI, and to also acquire worldwide aerospace STI of critical importance to NASA and the Nation. The program was established to support the objectives of NASA's missions and research and is dedicated to the advancement of aeronautics and space science. The NASA STI collection is available through its primary delivery platform—the NTRS (NASA Technical Reports Server).
Only NASA bulletins are used in this demo mode, but the program is scalable to any online pdf document.
APP Screenshots
