I built DocumentFlow as a standalone desktop app with Tauri + React, backed by AWS. The client runs in a native window, while the backend uses Python + FastAPI on Lambda, S3 for storage, MongoDB for data, OAuth2 + MFA for auth, and Docker for local dev.
Tauri + React Client
Wrapped my React UI in Tauri for a fast, native-feeling desktop app on Windows, macOS, and Linux.
AWS Backend & Security
Deployed FastAPI in Lambda, set up S3, IAM, CloudWatch logs, plus OAuth2 & MFA for rock-solid security.
Tech Stack
- Tauri
- React
- Python
- FastAPI
- AWS (Lambda, IAM, S3, CloudWatch)
- Docker
- OAuth2 & MFA
- MongoDB
Core Features I Built
Parsed uploads, preserved tables/lists, chunked and tagged.
Built a Neo4j graph to map entities and relationships.
Set up Pinecone for high-speed vector similarity searches.
Orchestrated multiple AI agents to handle parsing, QA, and output.
Refactored OpenAI's swarm library for production (for example adding async tool call support).
Edit files in real-time with AI (docx, pdf, powerpoint, txt).
How It Works
- Upload & Segment – Files are parsed, split, and entities are tagged.
- Index & Graph – Segments go to Pinecone; entities link up in Neo4j.
- Query & Retrieve – Agents vectorize queries and use metadata search on both detailed & summary indexes to fetch the most relevant chunks.
- Compile & Return – Responses are de-duplicated and shown in the UI chat.