-
-
Login page with JWT Authentication
-
Global AI Assistant for chatting with all the documents at once.
-
Settingspage including Profile and Account information
-
-
Page holding all the uploaded documents which can be opened and deleted.
-
-
For uploading Documents
-
AI Insights Page which shows document analysis, Executive summary, Storage summary, etc.
-
Dashboard page showing recent documents and system health.
-
Page that shows knowledge graph for all the uploaded documents if any relationships are found.
-
-
-
Page showing the opened document with pdf viewer and AI Assistant to chat with, about the particular document.
-
IndusBrain AI – Enterprise Knowledge Intelligence Platform
Inspiration
Modern enterprises generate thousands of documents such as manuals, SOPs, maintenance records, reports, and technical documentation. However, this valuable knowledge often remains scattered across multiple files and repositories, making it difficult for employees to quickly find accurate information. Traditional keyword-based search systems fail to understand context, leading to lost productivity and slower decision-making.
We built IndusBrain AI to transform static enterprise documents into an intelligent knowledge platform powered by Artificial Intelligence. Our goal was to enable users to search, understand, and interact with organizational knowledge using natural language while providing meaningful insights and visualizing relationships between information.
What It Does
IndusBrain AI is an AI-powered Enterprise Knowledge Intelligence Platform that enables organizations to:
- 📄 Upload and manage enterprise documents
- 🤖 Chat with documents using Retrieval-Augmented Generation (RAG)
- 🔍 Perform semantic AI-powered search across the knowledge base
- 🕸️ Visualize relationships using an interactive Knowledge Graph
- 📊 Generate AI-powered insights and analytics
- 👤 Secure user authentication and personalized workspaces
- 🌙 Support both Light and Dark modes for a modern user experience
Instead of searching through hundreds of pages manually, users can simply ask questions in natural language and receive accurate, context-aware answers.
How We Built It
The project follows a modern full-stack architecture.
Frontend
- Next.js (App Router)
- TypeScript
- Tailwind CSS
- shadcn/ui
Backend
- FastAPI
- Python
- SQLAlchemy
- SQLite
AI Stack
- Ollama (Qwen 2.5)
- Retrieval-Augmented Generation (RAG)
- ChromaDB
- Sentence Transformers
- OCR
- Entity Extraction
- Relationship Extraction
Security
- JWT Authentication
- HTTP-only Cookies
Key Features
- Secure Authentication
- AI Copilot
- Enterprise Search
- Knowledge Graph
- Dashboard Analytics
- AI Insights
- Document Management
- User Settings
- Dark Mode Support
Challenges We Faced
Developing an enterprise AI platform involved several technical challenges.
Efficient Document Retrieval
Retrieving the most relevant information from large collections of documents required building a robust semantic search pipeline using vector embeddings and Retrieval-Augmented Generation.
Knowledge Graph Generation
Extracting entities and relationships from unstructured documents and presenting them as an interactive knowledge graph required designing an efficient extraction workflow.
Full-Stack Integration
Connecting the Next.js frontend with the FastAPI backend while maintaining secure authentication and seamless communication required careful API design.
User Experience
Designing a clean, responsive, and enterprise-grade interface with consistent layouts, dark mode support, and intuitive navigation was equally important as implementing the AI functionality.
What We Learned
Throughout this project, we gained hands-on experience in:
- Building production-style AI applications
- Implementing Retrieval-Augmented Generation (RAG)
- Working with vector databases
- Designing scalable REST APIs using FastAPI
- Developing modern frontend applications with Next.js
- Managing authentication securely
- Creating intuitive enterprise dashboards
- Integrating AI models into real-world workflows
Beyond technical skills, we learned the importance of designing AI systems that solve real business problems while maintaining a seamless user experience.
Future Scope
We plan to enhance IndusBrain AI with several enterprise-grade capabilities, including:
- Multi-user collaboration
- Role-Based Access Control (RBAC)
- Multi-language document understanding
- Real-time document synchronization
- Voice-based AI assistant
- Cloud deployment
- AI-generated document summaries and reports
- Enterprise integrations with SharePoint, Google Drive, and Microsoft Teams
Impact
IndusBrain AI transforms enterprise knowledge into an intelligent, searchable, and interactive resource. By reducing the time spent searching through documents and enabling AI-assisted decision-making, it helps organizations improve productivity, preserve institutional knowledge, and make faster, data-driven decisions.
Built With
- ai
- chromadb
- fastapi
- machine-learning
- natural-language-processing
- next.js
- ocr
- ollama
- python
- rag
- react
- shadcn/ui
- sqlalchemy
- sqlite
Log in or sign up for Devpost to join the conversation.