Build a Multi-Agent AI Wedding Planner with LangChain | Full Project Step-by-Step

Build a real Multi-Agent Wedding Planner with LangChain, Python, OpenAI, Tavily, and Streamlit.

In this video, you'll learn how multi-agent AI systems actually work by building a practical portfolio-ready project step by step; not just theory, not just copy-paste code.

We'll build an AI wedding planning assistant that takes a couple's preferences, budget, location, guest count, style direction, constraints, planning priorities, and cultural details, then turns everything into a client-ready wedding plan.

The project uses a main wedding planner agent that coordinates two research subagents. Those subagents use Tavily web search to collect current venue, vendor, logistics, budget, and market information. Then the main agent synthesizes everything into a complete planning strategy with recommendations, timelines, risks, tradeoffs, and open questions.

By the end, you'll have a full LangChain portfolio project with both a Streamlit app and a CLI version.

🚀 What you'll learn
What multi-agent systems are and why they matter
How to build agents with LangChain
How to connect OpenAI models to LangChain
How to create a main agent that delegates work to subagents
How to expose subagents as tools
How to use Tavily search for live web research
How to structure prompts for a real AI planning workflow
How to build a Streamlit interface for an AI agent app
How to handle API keys with .env, Streamlit secrets, and sidebar inputs
How to save generated runs in Streamlit session state
How to export AI-generated output as Markdown
How to organize a clean AI backend portfolio project

🧠 Project features
Multi-Agent Wedding Planner built with LangChain
MainWeddingPlannerAgent for final planning and synthesis
SubAgent1 and SubAgent2 for delegated web research
Tavily search tool for current vendor, venue, logistics, and market data
OpenAI model integration through LangChain
Streamlit browser app with a polished planning workflow
CLI version for testing the agent flow from the terminal
Structured wedding brief form
Sample brief loader for quick demos
OpenAI and Tavily API-key readiness panel
Agent-client refresh button
Saved-run history
Generated plan tab
Input brief tab
Architecture tab
Markdown download button

🧩 What the final plan includes
Executive summary
Planning concept
Venue and vendor research recommendations
Budget allocation by category
Timeline from now through wedding day
Guest experience considerations
Risks and tradeoffs
Open questions for the couple

🏗️ Project architecture
Streamlit UI / CLI
- User wedding requirements
- Wedding planner prompt
- MainWeddingPlannerAgent
- delegate_to_subagent1(query)
- SubAgent1
- Tavily search_web(topic)
- delegate_to_subagent2(query)
- SubAgent2
- Tavily search_web(topic)
- Final synthesized wedding plan

🧠 Tech stack
Python 3.12+
LangChain
LangChain Agents
LangChain OpenAI
OpenAI
Tavily
Streamlit
dotenv
uv
VS Code
GitHub

⚡ Why this project matters
Most AI tutorials stop at one simple chatbot.

This project shows how to build a real agent workflow where one AI agent coordinates other agents, uses tools, performs live research, and produces a structured business-style output.

This is the kind of project you can add to your AI engineering portfolio because it demonstrates:
Agent orchestration
Tool calling
Multi-agent delegation
Prompt structuring
External API integration
Streamlit app development
CLI workflow design
Environment configuration
Practical AI product thinking

If you're learning LangChain, AI agents, backend engineering, or how to build real AI applications in Python, this project is a strong next step after your first beginner agent.

After this project, you can move into more advanced AI engineering topics like MCP, memory, state, context engineering, long-running conversations, human-in-the-loop workflows, specialized subagents, dynamic agent routing, and production-ready agent systems.

*Links below*
GitHub Repo: https://github.com/Mohamad-Hachem/MultiAgent_Wedding_Planner_With_Langchain
OpenAI Dashboard: https://platform.openai.com/home
LangChain: https://www.langchain.com/
Streamlit: https://streamlit.io/
Tavily: https://www.tavily.com/
uv: https://docs.astral.sh/uv/
Time for studying: https://pomofocus.io/

#AI #LangChain #Python #AIAgents #MultiAgent #LLM #OpenAI #Tavily #Streamlit #SoftwareEngineering #BackendEngineering #AIProjects #PromptEngineering #PortfolioProject #ArtificialIntelligence #PythonDeveloper #AgenticAI #AIAutomation