How nandakumar005/orian_ai works
Orian AI is a local browser-based RAG app for asking questions about office documents. A user selects a document folder and loads TXT, PDF, or DOCX files; the server routes that action to document loading, chunking, embedding, and persistent Chroma storage. For questions, the server opens the vector database and calls retrieval and answer generation, returning the response to the page. The README documents Ollama models and persistent local vectors. Edges follow the sampled app.py and rag.py calls or README workflow; no unsampled source behavior is assumed.
Main entry points: app.py Flask routes · config.py RAG settings · rag.py Similarity search · index.html Question form
Components 12
Browser interface
- Question form
templates/index.html - Flask routes
app.py
Document question answering
- Format loaders
rag.py - Text chunking
rag.py - Similarity search
rag.py - Context prompt
rag.py - RAG settings
config.py
Local models and storage
- Document embeddings
rag.py - Vector database
rag.py - Answer generation
rag.py
Other components
- User
- Office documents
Connections 18
- User → Question form: submits actions
README.md - Question form → Flask routes: sends requests
app.py - Flask routes → Format loaders: loads documents
app.py - Office documents → Format loaders: reads files
README.md - Format loaders → Text chunking: splits text
rag.py - Text chunking → Document embeddings: creates vectors
README.md - Document embeddings → Vector database: stores vectors
rag.py - Flask routes → Vector database: opens database
app.py - Flask routes → Similarity search: asks question
app.py - Similarity search → Vector database: searches vectors
rag.py - Similarity search → Context prompt: builds context
rag.py - Context prompt → Answer generation: generates answer
rag.py - Answer generation → Flask routes: returns response
app.py - RAG settings → Text chunking: sets chunking
rag.py - RAG settings → Document embeddings: sets model
rag.py - RAG settings → Similarity search: sets search
rag.py - RAG settings → Answer generation: sets model
rag.py - Vector database → User: persists locally
README.md