PromptCompiler 🚀

AI-Powered Prompt Engineering Platform
Transform vague intent into production-ready prompts through intelligent compilation

Live Demo • Documentation • Architecture

---

📖 Overview

PromptCompiler is a production SaaS platform that bridges the gap between user intent and professional-grade AI prompts. Using an 8-stage AI compilation pipeline powered by multi-agent orchestration, it transforms natural language inputs into structured, platform-optimized prompts with contextual awareness, role definitions, and domain-specific constraints.

🎯 Core Value Proposition

• Democratizes Prompt Engineering: Enables non-technical users to create expert-level prompts through natural language
• AI-Powered Optimization: Uses LangGraph multi-agent orchestration for semantic analysis, strategy planning, and workflow synthesis
• Platform-Aware Generation: Tailored outputs for Instagram, YouTube, Blogs with format-specific constraints
• Scalable SaaS Architecture: Tiered subscription model with usage tracking and authentication

---

✨ Key Features

🧠 8-Stage Compilation Pipeline

User Intent → Semantic Analysis → Strategy Selection → Workflow Planning
    ↓
Agent Orchestration → Parallel Execution → Result Synthesis → Metadata Enrichment
    ↓
Structured Prompt Output (Markdown)

• Intent Classification: Natural language understanding of user goals
• Semantic Analysis: Extract key requirements, constraints, and context
• Strategy Engine: Select optimal compilation workflow based on domain
• Agent Orchestration: Multi-agent system using LangGraph state graphs
• Dynamic Execution: Conditional branching and parallel agent execution
• Result Synthesis: Manager agent consolidates outputs into cohesive prompts
• Metadata Tracking: Latency, strategy used, and credit consumption

🎨 Domain-Specific Workflows

✅ Content Creation (Live)

• Instagram Posts: Character limits, hashtag strategies, CTAs
• YouTube Scripts: Hook-body-closure structure, retention optimization
• Blog Articles: SEO best practices, heading structure, readability

🚧 Coming Soon

• Developer Prompts: Code generation, documentation, system design
• Research & Analysis: Deterministic reasoning chains, synthesis workflows

🔐 Authentication & Security

• Supabase Auth Integration: JWT/JWKS verification
• Auto-Profile Creation: Seamless onboarding with metadata extraction
• Row-Level Security (RLS): Database isolation per user
• Protected API Endpoints: All compilation requests require authentication

📊 Usage Tracking & Subscriptions

Tier	Monthly Limit	Features	Price
Free	5 prompts	Basic platforms, standard AI	$0
Pro	Coming	All platforms, priority support	$9.99
Enterprise	coming	API access, custom models, SSO	Custom

• Atomic Usage Counting: Transaction-safe increments
• Monthly Limit Enforcement: Automated plan-based restrictions
• Conversion Tracking: Free-to-paid upgrade analytics

---

🛠️ Technical Architecture

System Design

┌─────────────┐      ┌──────────────┐      ┌─────────────┐
│   Next.js   │──────│   FastAPI    │──────│   Groq API  │
│  Frontend   │ HTTP │   Backend    │ REST │  (Llama-3)  │
└─────────────┘      └──────────────┘      └─────────────┘
       │                     │                      │
       │                     │                      │
       ▼                     ▼                      ▼
┌─────────────┐      ┌──────────────┐      ┌─────────────┐
│  Supabase   │◀─────│  LangGraph   │      │  LangSmith  │
│    Auth     │      │ Orchestrator │      │  (optional) │
└─────────────┘      └──────────────┘      └─────────────┘
       │                     │
       ▼                     ▼
┌──────────────────────────────────────┐
│      Supabase (PostgreSQL)           │
│  ├─ User Profiles                    │
│  ├─ Usage Tracking                   │
│  └─ Compilation History              │
└──────────────────────────────────────┘

Tech Stack

Backend

• Runtime: Python 3.9+ with FastAPI 0.128.0
• AI Framework: LangGraph 1.0.7 + LangChain Core 1.2.7
• LLM Provider: Groq API with Llama-3-70B-8192
• Database: Supabase (PostgreSQL) with RLS
• Server: Uvicorn with async support
• Deployment: Render (cloud hosting)

Frontend

• Framework: Next.js 16.1.6 (App Router)
• Language: TypeScript 5.x
• Auth: Supabase SSR (@supabase/ssr 0.8.0)
• UI: Radix UI + Tailwind CSS 4.0
• Animations: Framer Motion 12.29.2
• State Management: React Context API
• Deployment: Vercel

---

📋 API Endpoints

Public Endpoints

Method	Endpoint	Description
GET	/health	Service health check

Protected Endpoints (Require JWT)

Method	Endpoint	Description	Request Body
GET	/me	User profile + usage stats	-
POST	/compile	Generate optimized prompt	{ "intent": string, "domain": string, "platform?": string }
GET	/history	Compilation history	-

Example Request

curl -X POST https://your-backend.onrender.com/compile \
  -H "Authorization: Bearer YOUR_JWT_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "intent": "Create a YouTube video script for beginners about Python",
    "domain": "content_creation",
    "platform": "youtube"
  }'

Example Response

{
  "prompt": "# Role\nYou are an expert YouTube script writer...",
  "metadata": {
    "strategy": "content_creation_youtube",
    "latency_ms": 2341,
    "credits_used": 1,
    "timestamp": "2026-01-15T10:30:00Z"
  }
}

---

🏗️ Project Structure

Backend (/)

promptcompiler-backend/
├── app/
│   ├── main.py                 # FastAPI application entry
│   ├── api/                    # API route handlers
│   │   ├── compile.py          # /compile endpoint
│   │   ├── profile.py          # /me endpoint
│   │   └── history.py          # /history endpoint
│   ├── auth/                   # Authentication logic
│   │   ├── jwt_verifier.py     # Supabase JWT verification
│   │   └── dependencies.py     # Auth dependencies
│   ├── core/                   # Core compilation engine
│   │   ├── compiler.py         # Main compilation orchestrator
│   │   ├── intent/             # Intent classification
│   │   ├── semantic/           # Semantic analysis
│   │   ├── strategy/           # Strategy selection
│   │   ├── workflow/           # Workflow planning
│   │   ├── agents/             # Individual agent implementations
│   │   └── decisions/          # Decision-making logic
│   ├── orchestration/          # LangGraph state management
│   │   ├── graph.py            # State graph definitions
│   │   └── nodes.py            # Graph node implementations
│   ├── service/                # Business logic services
│   │   ├── usage_service.py    # Usage tracking
│   │   └── profile_service.py  # Profile management
│   ├── models/                 # Pydantic data models
│   │   ├── request.py          # Request schemas
│   │   └── response.py         # Response schemas
│   └── Database/               # Database utilities
├── tests/                      # Comprehensive test suite
├── requirements.txt            # Python dependencies
└── Procfile                    # Deployment config (Render)

Frontend (/apps/frontend)

apps/frontend/
├── app/                        # Next.js App Router
│   ├── page.tsx                # Landing page
│   ├── layout.tsx              # Root layout
│   ├── compiler/               # Compilation interface
│   ├── auth/                   # Login/signup pages
│   ├── profile/                # User profile
│   ├── history/                # Compilation history
│   └── pricing/                # Pricing page
├── components/                 # Reusable UI components
│   ├── ui/                     # Shadcn/Radix components
│   ├── CompilerForm.tsx        # Main form component
│   ├── ResultDisplay.tsx       # Prompt output viewer
│   └── DomainSelector.tsx      # Platform selection
├── context/                    # React Context providers
│   └── CompilerContext.tsx     # Global state management
├── lib/                        # Utility functions
│   ├── supabase.ts             # Supabase client
│   └── api.ts                  # Backend API client
├── public/                     # Static assets
└── package.json                # Node dependencies

---

Production URL: https://your-service.onrender.com

🔮 Roadmap

✅ Completed (v1.0)

• 8-stage compilation pipeline
• Multi-agent LangGraph orchestration
• Supabase authentication & RLS
• Usage tracking & tiered limits
• Content creation workflows (Instagram, YouTube, Blogs)
• Production-ready FastAPI backend
• Next.js frontend with SSR auth
• Vercel + Render deployment

🚧 In Progress (v1.1)

• Compilation history persistence
• Result sharing (public URLs)
• Copy-to-clipboard enhancement
• Mobile navigation improvements

🔜 Planned (v2.0)

• Developer prompt workflows (code, docs, system design)
• Research & analysis workflows (reasoning chains)
• Template library (pre-built prompts)
• Team collaboration features
• Advanced analytics dashboard
• API rate limiting (Redis)
• Payment integration (Stripe)
• Custom AI model configuration

🌟 Future Vision (v3.0+)

• SSO integration (SAML/OAuth)
• Public API platform
• Webhook support
• White-label deployments
• Multi-language support
• Plugin marketplace

---

• Live Demo: https://prompt-compiler-frontend.vercel.app/

---

🙏 Acknowledgments

• LangChain & LangGraph: For the multi-agent orchestration framework
• Groq: For ultra-fast LLM inference
• Supabase: For auth, database, and RLS
• FastAPI: For the modern Python web framework
• Next.js: For the React SSR framework
• Vercel & Render: For seamless deployment

---

📊 Project Stats

• Backend Lines: ~3,000 (Python)
• Frontend Lines: ~2,500 (TypeScript/TSX)
• Test Coverage: 85%+
• API Response Time: < 2s (95th percentile)
• Uptime: 99.9% SLA
• Active Users: Growing via freemium model

---

Built with ❤️ by Aman Kushwaha

⬆ Back to Top