InstagramClone_withFastAPI

Simple Social is a lightweight Instagram-inspired playground that pairs a FastAPI backend with a Streamlit frontend. Users can register/login with JWT auth, upload images or short videos to ImageKit, and browse/delete their own posts from a unified feed.

Tech Stack

• FastAPI + FastAPI Users for auth, dependency injection, and JWT handling
• Async SQLAlchemy + SQLite for persistent storage
• ImageKit for hosted media, resizing, and caption overlays
• Streamlit for a fast, password-protected web UI
• uv + pyproject.toml for reproducible Python environments

Getting Started

1. Prerequisites

• Python 3.11+
• uv (recommended) or a virtualenv tool of your choice
• ImageKit account (for API keys)

2. Environment variables

Create a .env file in the project root with your ImageKit credentials:

IMAGEKIT_PUBLIC_KEY=pk_xxx
IMAGEKIT_PRIVATE_KEY=sk_xxx
IMAGEKIT_URL=https://ik.imagekit.io/<your_id>

3. Install dependencies

cd "/Users/vishnuanand/Documents/Purdue Files/Instagram_Clone_FastAPI/InstagramClone_withFastAPI"
uv sync

If you prefer pip, create and activate a virtualenv and run pip install -e ..

4. Run the backend API

uv run python main.py
# or
uv run uvicorn app.app:app --reload --host 0.0.0.0 --port 8000

The SQLite database (test.db) and tables are created automatically on startup.

5. Launch the Streamlit client

In a second terminal (while the API keeps running):

uv run streamlit run frontend.py

Visit the URL Streamlit prints (typically http://localhost:8501). Register a user, log in, and start posting.

Topics Explored While Building

• Structuring a FastAPI app with a lifespan context to bootstrap async DB connections.
• Using FastAPI dependency injection to share the async SQLAlchemy session and enforce current_active_user.
• Integrating FastAPI Users to avoid hand-rolling JWT auth, password hashing, and reset/verify flows.
• Designing async SQLAlchemy models with UUID primary keys and relationships between User and Post.
• Handling media uploads with UploadFile, temp files, and piping them to ImageKit, including text overlays for captions.
• Building a simple but stateful Streamlit UI that handles auth tokens, feeds, uploads, and deletions via REST calls.

What I Learned

• How JWT auth works end-to-end when FastAPI Users handles token issuance while Streamlit stores tokens in st.session_state.
• The ergonomics of async SQLAlchemy sessions—committing, refreshing, and querying without blocking the event loop.
• Practical secrets management using .env + python-dotenv to keep API keys out of the codebase.
• The tradeoffs of using hosted media storage (ImageKit) versus storing blobs locally, especially for transformations.
• How quickly Streamlit can serve as a full-featured client for a REST API, making backend testing far more visual.

Feel free to open issues or extend the project—ideas include comments, likes, follower graphs, and richer notification flows.