Personalized date planning powered by Groq AI and real-time data
Helping couples create memorable experiences through intelligent planning
Planning a date involves juggling multiple factors: personality compatibility, budget constraints, venue quality, timing, and local context. Most people spend hours researching venues, reading reviews, and trying to match their partner's preferences — often resulting in generic, uninspired plans.
The problem gets worse when:
- You are in an unfamiliar city
- You have specific dietary or lifestyle constraints
- You want to surprise your partner with something thoughtful
- You are working with a tight budget
AI Date Planner is an intelligent system that generates personalized, contextual date plans in under 15 seconds. It combines:
- Personality-driven matching using visual chip selectors for intuitive input
- Real-time venue search with quality filtering (4.0+ rating, 50+ reviews)
- RAG-powered recommendations from curated knowledge base documents
- Budget optimization ensuring plans stay within your range
- Streaming AI generation with live progress updates
The system orchestrates multiple data sources in parallel, applies intelligent filtering, and crafts a complete date narrative — all while respecting cultural context and safety considerations for Indian cities.
| Feature | Description |
|---|---|
| Personality Matching | Select from 20 personality traits, 20 interests, and 10 dislikes using visual chips |
| Quality Filtering | Only venues with 4.0+ ratings and 50+ reviews; excludes inappropriate types |
| Parallel Execution | Tools run concurrently, reducing generation time from 30s to 10-15s |
| Real-time Streaming | Watch your date plan appear progressively with step-by-step status updates |
| City-Specific Context | Curated guides covering Jaipur, Delhi, Mumbai, Bangalore with budget tiers |
| Budget Intelligence | Automatic cost estimation and validation across all segments |
| Gift & Flower Recommendations | Personalized suggestions based on occasion and personality |
| Google Maps Integration | Direct links to venues with addresses and ratings |
User Input -> Preference Vector -> Parallel Tool Execution -> RAG Retrieval -> AI Generation -> Structured Plan
- Profile Creation: Users select personality traits, interests, and dislikes using visual chips
- Location Detection: Browser geolocation provides current coordinates
- Parallel Tool Execution: Three tools run simultaneously via
asyncio.gather():- Places API searches nearby venues with quality filters
- Gift engine generates personalized recommendations
- Flower selector matches bouquets to occasion
- RAG Retrieval: Query embeddings (sentence-transformers) fetch relevant city guides and dating tips
- AI Generation: Groq (llama-3.3-70b-versatile) streams a narrative plan in real-time
- Budget Validation: Final plan is checked against budget constraints
+-------------------------------------------------------------+
| Frontend (React + Vite) |
| +--------------+ +--------------+ +--------------+ |
| | ChipSelector | | LoadingState | | DatePlanView | |
| +--------------+ +--------------+ +--------------+ |
| | |
| SSE Streaming |
+-------------------------+------------------------------------+
|
+-------------------------+------------------------------------+
| Backend (Python / Flask) |
| |
| +------------------+------------------+ |
| | Planner Agent (Async) | |
| +------------------+------------------+ |
| | |
| +------------------+------------------+ |
| | | | |
| +----+----+ +------+-----+ +-------+-----+ |
| | Places | | Gifts | | Flowers | |
| | Tool | | Tool | | Tool | |
| +----+----+ +------+-----+ +-------+-----+ |
| | | | |
| +------------------+------------------+ |
| asyncio.gather() |
| | |
| +------------------+------------------+ |
| | RAG System | |
| | +------------+ +------------+ | |
| | | Embeddings | | Retrieval | | |
| | +------------+ +------------+ | |
| +------------------+------------------+ |
+-------------------------------------------------------------+
|
+-------------------------------------------------------------+
| External Services |
| +--------------+ +--------------+ |
| | Groq AI | | Google Places| |
| | llama-3.3-70b| | API | |
| +--------------+ +--------------+ |
+-------------------------------------------------------------+
| Step | User Action | System Response |
|---|---|---|
| 1 | Opens app | Displays multi-step form with chip selectors |
| 2 | Selects personality traits (e.g., Romantic, Foodie) | Chips highlight on selection |
| 3 | Chooses interests and dislikes | Max limits enforced, summary shown |
| 4 | Enters budget range (Rs.1000-3000) | Validates min/max input |
| 5 | Allows location access | Browser geolocation captured |
| 6 | Submits form | Streaming begins: "Finding venues..." |
| 7 | Waits 10-15 seconds | Progress steps update: venues -> gifts -> flowers -> crafting |
| 8 | Watches text stream | Date plan appears progressively with blinking cursor |
| 9 | Reviews complete plan | Sees timeline, venues, gifts, flowers, budget breakdown |
| 10 | Clicks venue links | Opens Google Maps for navigation |
- React 18 with TypeScript for type safety
- Vite for fast development and optimized builds
- Vanilla CSS with custom design system (Playfair Display + DM Sans fonts)
- Server-Sent Events (SSE) for real-time streaming
- Python 3.9+ with Flask for REST API
- asyncio + aiohttp for parallel tool execution
- Groq SDK for LLM integration (llama-3.3-70b-versatile)
- sentence-transformers (all-MiniLM-L6-v2) for local embeddings
- NumPy for vector operations in RAG system
- Groq (llama-3.3-70b-versatile) for fast text generation and streaming
- sentence-transformers for semantic search embeddings (runs locally, no API needed)
- Google Places API for real-time venue data
- RAG System with curated knowledge base documents (city guides, dating tips)
- Complete frontend with chip selectors and streaming UI
- Async backend with parallel tool execution
- RAG system with knowledge base documents
- Venue quality filtering (4.0+ rating, 50+ reviews)
- Real-time streaming with progress indicators
- Budget validation and cost estimation
- Gift and flower recommendation engines
- Google Maps integration
- User authentication and profile persistence
- Saved date plans and history
- Vendor partnerships for bookings
- Mobile app (React Native)
- Multi-language support
- Python 3.9 or higher
- Node.js 16 or higher
- A Groq API key (free)
- A Google Places API key
cd backend
# Install dependencies
pip install -r requirements.txt
# Copy the env template and fill in your keys
cp .env.example .env
# Edit .env with your GROQ_API_KEY and GOOGLE_PLACES_API_KEY
# Run the server
python app.pyBackend starts at http://localhost:5001
cd web
# Install dependencies
npm install
# Start development server
npm run devFrontend starts at http://localhost:3000
- Groq API Key: Visit console.groq.com — free tier available
- Google Places API Key: Visit Google Cloud Console
- Enable the Places API for your project
- Create credentials -> API Key
AI-Date-Planner/
+-- backend/
| +-- planner/
| | +-- planner_agent.py # Orchestrates the full date planning pipeline
| +-- rag/
| | +-- city_guides.json # Curated city-specific dating guides
| | +-- embed.py # Embedding + knowledge base initialization
| | +-- kb_docs.py # Static knowledge base documents
| | +-- retrieve.py # Cosine similarity retrieval
| +-- tools/
| | +-- budget_tool.py # Budget validation and suggestions
| | +-- flowers_tool_async.py # Flower recommendations by occasion
| | +-- gifts_tool_async.py # AI-generated gift ideas via Groq
| | +-- places_tool_async.py # Google Places venue search
| +-- app.py # Flask entry point and API routes
| +-- config.py # Environment config
| +-- groq_client.py # Groq LLM + embedding client
| +-- models.py # Python dataclasses (request/response models)
| +-- requirements.txt
| +-- .env.example # Copy this to .env
+-- web/
| +-- src/
| | +-- components/
| | | +-- ChipSelector.tsx # Reusable pill-chip multi-select
| | | +-- DateForm.tsx # Multi-step form (3 steps)
| | | +-- DatePlanView.tsx # Renders the completed date plan
| | | +-- LoadingState.tsx # Animated progress + streaming preview
| | +-- App.tsx # Root component + SSE streaming logic
| | +-- index.css # All styles and design tokens
| | +-- main.tsx # React entry point
| | +-- types.ts # TypeScript interfaces
| +-- index.html
| +-- vite.config.ts # Vite + proxy config (port 3000 -> 5001)
+-- .gitignore
+-- README.md
We welcome contributions! If you would like to help improve the AI Date Planner, please ⭐ Star and 🍴 Fork the repository, then submit a pull request with your changes.
- Additional city guides for the knowledge base
- UI/UX improvements
- Unit and integration test coverage
- Performance optimization
This project is licensed under the MIT License.
- Groq for fast, free LLM inference
- Google Places API for real-time venue data
- Hugging Face / sentence-transformers for local embedding models
- React and Vite community for excellent tooling
Built for couples who deserve better date nights.