PratibimbAI Backend 🚀

PratibimbAI is a production-grade Agentic AI system designed for professional content creators to mimic their voice and automate high-quality social media posts.

🏗 Architecture

The system is built on a domain-driven architecture for maximum scalability and contributor friendliness.

graph TD
    A[API Layer] --> B[Orchestration Service]
    B --> C[Redis Queue]
    C --> D[Worker]
    D --> E[LangGraph Agent]
    
    subgraph Agent Flow
        E --> F[Collect]
        F --> G[Clean]
        G --> H[Rank]
        H --> I[Hook]
        I --> J[Write]
        J --> K[Critic]
        K --> L[Verify]
        L -- Retry --> J
    end
    
    subgraph Services
        F --- Ingestion[Browser/Static Scrapers]
        H --- Processing[Embeddings/Ranker]
        J --- LLM[OpenAI/Gemini/Groq]
    end
    
    subgraph Data
        DataLayer[Database Layer] --- Supabase[(PostgreSQL)]
        CacheLayer[Cache Layer] --- Redis[(Redis)]
    end

Loading

🌟 Key Features

• Ghostwriter Mode: Learns your specific writing style from samples using RAG.
• Distributed Jobs: Redis + RQ for scalable, asynchronous processing.
• Fault-Tolerant Agents: Node-level retry wrappers and verification loops.
• Semantic Caching: Avoids redundant scraping of the same URLs.
• Structured Prompts: Categorized prompts for generation, analysis, and ideation.
• Multi-LLM Support: Cost-optimized tiering across Groq, Gemini, and OpenAI.

🛠 Project Structure

• app/api/: FastAPI routes and schemas.
• app/graph/: LangGraph definitions, nodes, and retry wrappers.
• app/services/: Domain-driven services (Ingestion, LLM, Processing, Orchestration).
• app/jobs/: Job lifecycle management (Service, Repository, Worker).
• app/prompts/: Categorized prompt templates.
• app/utils/: Shared utilities (Tracing, Metrics, Cache, Auth).

🚀 Getting Started

Local Development

1. Clone the repo
2. Setup Virtual Env

   cd backend
   python -m venv .venv
   source .venv/bin/activate
   pip install -r requirements.txt
   playwright install chromium

3. Configure Environment Copy .env.example to .env and fill in your keys.
4. Run Stack

   docker run -p 6379:6379 redis
   rq worker pratibimb_jobs
   uvicorn app.main:app --reload

📜 Documentation

Detailed documentation can be found in the docs/ folder:

• Architecture Overview
• Agent Workflow
• API Reference

🤝 Contributing

We welcome contributions! Please see our Contributing Guide to get started.