Invantia

About Invantia

Intelligent Corpus Reduction for Large Document Sets

🏢 Company

Invantia is a web property of TrxSys LLC, a technology research and development company focused on document intelligence and AI-assisted analysis.

For more information about TrxSys LLC, visit trxsys.com

📦 Products

Available Now Invantia Desktop

Free, privacy-first document intelligence platform that runs entirely in your browser.

  • 100% client-side processing - your documents never leave your device
  • Semantic search with vector embeddings for intelligent content discovery
  • Hybrid scoring combining exact matching and semantic expansion
  • Super-chunking approach leveraging large LLM context windows
  • Generate "chat packages" for use with ChatGPT, Claude, Gemini, or any LLM
  • IndexedDB storage - no server uploads, no cloud dependencies
  • Collections for organizing related document sets
  • Supports PDF, DOCX, and TXT formats

Perfect for individuals and small teams who need powerful document analysis while maintaining complete data privacy.

Try Desktop Now

Under Development Invantia Enterprise

Server-side document intelligence for organizations requiring centralized processing, advanced workflows, and team collaboration.

  • Centralized document repository with role-based access control
  • Advanced query orchestration and workflow automation
  • Composable primitives for building complex analysis pipelines
  • Multi-user collections and shared query templates
  • Audit trails and execution chain transparency
  • API access for integration with existing systems
  • Priority support and custom deployment options

Designed for legal firms, accounting practices, research organizations, and enterprises handling large document volumes requiring sophisticated analysis.

Enterprise Edition is currently in development. Contact us for early access information.

⚖️ Legal & Licensing

Open Source

Invantia's source code is available on GitHub under an open source license.

View on GitHub

Copyright & Rights

Copyright © 2024-2025 TrxSys LLC. All rights reserved.

While the source code is open source, the Invantia name, branding, and associated materials are proprietary to TrxSys LLC.

Privacy & Security

Invantia Desktop operates in privacy-first mode by default:

  • No cookies - Desktop edition does not use cookies for tracking
  • No analytics - No Google Analytics, no third-party tracking scripts
  • No data transmission - Documents and queries are never sent to our servers
  • Local storage only - All data stored in browser IndexedDB
  • No external dependencies - Operates completely offline after initial page load
  • User-controlled LLM integration - You choose which AI service to use

For comprehensive security and privacy information, see our Information Security page.

Disclaimer

Invantia is a tool for document analysis and information retrieval. It does not provide legal, financial, medical, or professional advice. Users are responsible for verifying the accuracy of any information processed through the platform and for ensuring compliance with applicable laws and regulations in their jurisdiction.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY.

👤 About the Founder

Robert Lawson Vaughn

Portland, Oregon

Robert Lawson Vaughn is a technology researcher and inventor with over 75 technical patents spanning artificial intelligence, document processing, security systems, and enterprise software architecture.

Currently serving as AI Domain Owner at Intel IT, Robert brings decades of experience in enterprise-scale technology development to Invantia's design and architecture.

His work focuses on bridging the gap between cutting-edge AI capabilities and practical, privacy-conscious tools that professionals can trust with sensitive documents. Invantia represents his vision for document intelligence that respects user privacy while delivering sophisticated analytical capabilities.

GitHub Profile

🔧 Technology & Architecture

Core Philosophy: Intelligent Corpus Reduction

Traditional RAG (Retrieval-Augmented Generation) systems face a fundamental challenge: context window limitations force compromising between completeness and relevance.

Invantia takes a different approach: intelligent corpus reduction. Rather than fragmenting documents into tiny chunks and hoping the right pieces are retrieved, Invantia uses user-guided queries to progressively reduce large document sets down to precisely the content needed for comprehensive LLM analysis.

This "super-chunking" approach leverages modern LLMs' large context windows (100K+ tokens) to preserve co-reference relationships and complex document structures that traditional RAG systems fragment.

Desktop Architecture

  • Frontend: Pure JavaScript, HTML5, CSS3
  • Storage: IndexedDB for local document persistence
  • Processing: Client-side PDF.js, Mammoth.js for document parsing
  • Search: TF-IDF vectorization with semantic expansion
  • Privacy: Zero server communication after initial page load

Enterprise Architecture (In Development)

  • Backend: FastAPI (Python) with async processing
  • Database: SQLite for metadata, file system for document storage
  • Processing: Server-side PDF extraction, OCR capabilities
  • API: RESTful endpoints for integration
  • Deployment: Self-hosted, behind nginx reverse proxy

📧 Contact & Support

For questions, feature requests, or support inquiries: