Skip to content

PiyushWithPant/ScholarGPT

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

📚 ScholarGPT

ScholarGPT is a production-ready, zero-cost Retrieval-Augmented Generation (RAG) workspace designed to extract precise insights from your personal & private research data. By leveraging LangChain, Streamlit, and an in-memory ChromaDB vector store, the application creates an isolated semantic search matrix for any uploaded PDF, allowing users to execute deterministic Q&A without data persistence or cloud-hosting overhead so your data remains safe completely.


🛠️ Tech Stack & Architecture

  • Orchestration: LangChain Expression Language (LCEL)
  • Vector Database: ChromaDB (Transient, running entirely in-memory per user session)
  • Embeddings Model: Google AI Studio gemini-embedding-001
  • Generative LLM: Groq's llama-3.3-70b-versatile (Configured at temperature=0 for anti-hallucination guardrails)

🚀 Key Features

  • Zero-Cost Infrastructure: Runs entirely inside free developer tiers and local server memory.
  • Transient Data Lifecycle: Uploaded documents are parsed, chunked, and indexed on-the-fly. No data is written to permanent storage, ensuring absolute session privacy.
  • Recursive Parsing: Leverages a RecursiveCharacterTextSplitter configured for a 1,000-character chunk size and 200-character overlap to preserve semantic context across multi-column academic layouts.
  • Rate-Limit Safeguards: Implements Streamlit form barriers to freeze execution during active typing, preventing rapid keystroke script refreshes from exhausting API quotas.

📂 Project Structure

scholar_gpt/
│
├── .env                # Local API Keys (Excluded from version control)
├── .gitignore          # Git exclusion rules 
├── requirements.txt    # Frozen python package dependencies
├── backend.py          # Vector pipeline, LCEL chains, and ingestion logic
└── app.py              # Reactive UI layout and state management engine

⚙️ Local Installation & Setup

  1. Clone the Workspace
git clone https://github.com/PiyushWithPant/ScholarGPT.git

Go to the root dir

cd scholar-gpt
  1. Create virtual env
python -m venv env
  1. Activate env
  • On MacOS
source env/bin/activate
  • On Windows
env\Scripts\activate
  1. Install all dependencies
pip install -r requirements.txt
  1. Configure Environment Secrets
GOOGLE_API_KEY="YourSecretKey..."
  1. Launch the streamlit server
streamlit run app.py

⚖️ License

Distributed under the MIT License. See LICENSE for more information.


By Piyush Pant (पियुष पंत)

About

Your Personal AI Research Assistant & Study Buddy.

Topics

Resources

License

Stars

1 star

Watchers

0 watching

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages