{"id":"903f77f6-2bf6-4eac-b0e3-fb1811557c69","entityType":"agent","slug":"crewai-ashwin14101-langchain-langgraph-agents","name":"langchain-langgraph-agents","canonicalUrl":"https://www.xpersona.co/agent/crewai-ashwin14101-langchain-langgraph-agents","canonicalPath":"/agent/crewai-ashwin14101-langchain-langgraph-agents","generatedAt":"2026-10-09T00:03:11.433Z","source":"GITHUB_OPENCLEW","claimStatus":"UNCLAIMED","verificationTier":"NONE","summary":{"evidence":{"source":"editorial-content","verified":true,"confidence":"high","updatedAt":"2026-05-31T06:18:26.077Z","emptyReason":null},"description":"Structured collection of LangChain & LangGraph notebooks: prompt chains, RAG, ReAct agents, stateful graphs with MemorySaver, multi-agent CrewAI systems, and a career AI agent. Built with Gemini + Tavily. 🤖 LangChain & LangGraph — Agents, RAG & Multi-Agent Systems

\n \n \n \n \n \n \n

\n\n

\n A structured collection of notebooks covering the full LangChain and LangGraph ecosystem —\n from prompt templates and chains to stateful multi-agent systems, RAG pipelines, and CrewAI orchestration.\n

\n\n---\n\n## 📂 Notebooks\n\n| Notebook | Topics Covered | Key Concepts |\n|---|---|---|\n| [`Langchain.ipynb`](./Langchain.ipynb) | LangChain foundations | Prompt templates, chains, LCEL, structured output, Pydantic, parsers |\n| [`LLM_Document_Loaders_And_RAG.ipynb`](./LLM_Document_Loaders_And_RAG.ipynb) | Document loaders + splitters | TXT, PDF, CSV, Web loaders; Character, Recursive, Code splitters; Chroma |\n| [`Rag_Implementation.ipynb`](./Rag_Implementation.ipynb) | RAG from scratch | PDF loader, text splitting, Chroma vector store, RetrievalQA |\n| [`Langchain_Agents.ipynb`](./Langchain_Agents.ipynb) | LangChain agents | Tool use, Tavily search, custom tools with `@tool`, ReAct agents |\n| [`Langgraph_Agents.ipynb`](./Langgraph_Agents.ipynb) | LangGraph agents | MessageGraph, StateGraph, planner-search-responder pipeline |\n| [`Chatbot_Graph.ipynb`](./Chatbot_Graph.ipynb) | Stateful chatbot | MemorySaver, multi-turn conversation, thread-based memory |\n| [`Post_Creation_Agent.ipynb`](./Post_Creation_Agent.ipynb) | Multi-node graph | Writer → Reviewer agent loop, LinkedIn post generation |\n| [`Crew.ipynb`](./Crew.ipynb) | CrewAI multi-agent | Crew, Task, Agent orchestration, role-based AI agents |\n| [`Career_Agent.ipynb`](./Career_Agent.ipynb) | Career AI agent | Resume + job analysis agent with Tavily real-time search |\n\n---\n\n## 🧠 Concepts Covered\n\n### 1. LangChain Core (`Langchain.ipynb`)\n- **Prompt Templates**: `ChatPromptTemplate`, `PromptTemplate`, multi-message templates\n- **LCEL (LangChain Expression Language)**: `prompt | model | parser` chains\n- **Structured Output**: `with_structured_output()` with TypedDict, Pydantic, and JSON schema\n- **Output Parsers**: `StrOutputParser`, `JsonOutputParser`, custom parsers\n- **Multi-model**: Gemini, Mistral (via OpenRouter), OpenAI-compatible APIs\n\n### 2. RAG Pipeline (`Rag_Implementation.ipynb`)\n```\nPDF → PyPDFLoader → RecursiveCharacterTextSplitter (chunk_size=1000, overlap=50)\n → HuggingFace Embeddings → Chroma VectorStore → RetrievalQA (k=8)\n```\n\n### 3. Agents & Tools (`Langchain_Agents.ipynb`)\n- `initialize_agent` with `zero-shot-react-description`\n- `TavilySearchResults` for real-time web search\n- Custom tools with `@tool` decorator\n- Tool selection and reasoning trace\n\n### 4. LangGraph (`Langgraph_Agents.ipynb`, `Chatbot_Graph.ipynb`)\n```\nStateGraph / MessageGraph\n ├── Planner Node → decides what to search\n ├── Search Node → calls Tavily\n └── Responder Node → generates final answer\n```\n- `MemorySaver` for persistent multi-turn memory\n- `thread_id` for session isolation\n- `add_messages` annotation for message accumulation\n\n### 5. Multi-Agent Collaboration (`Crew.ipynb`)\n- **CrewAI**: `Agent`, `Task`, `Crew` orchestration\n- Role-based agents with backstory and goal definitions\n- Multi-agent task delegation and output chaining\n\n### 6. Post Creation Agent (`Post_Creation_Agent.ipynb`)\n- Writer Node → writes LinkedIn post\n- Reviewer Node → critiques and improves\n- Cyclic graph with conditional edge back to writer\n\n---\n\n## 🚀 Setup\n\n### 1. Clone\n\n```bash\ngit clone https://github.com/Ashwin14101/langchain-langgraph-agents.git\ncd langchain-langgraph-agents\n```\n\n### 2. Install dependencies\n\n```bash\npip install langchain langchain-core langchain-community langchain-google-genai \\\n langgraph crewai tavily-python chromadb sentence-transformers \\\n pypdf python-dotenv\n```\n\n### 3. Set API keys\n\nCreate a `.env` file:\n```env\nGOOGLE_API_KEY=your_google_api_key_here\nTAVILY_API_KEY=your_tavily_api_key_here\n```\n\n> **Get your keys:**\n> - Google AI Studio → https://aistudio.google.com/app/apikey (free)\n> - Tavily → https://tavily.com (free tier available)\n\n### 4. In each notebook, replace the API key cell with:\n\n```python\nfrom dotenv import load_dotenv\nimport os\nload_dotenv()\nAPI_KEY = os.getenv(\"GOOGLE_API_KEY\")\n```\n\n### 5. Run notebooks\n\n```bash\njupyter notebook\n```\n\n---\n\n## 🛠️ Tech Stack\n\n| Tool | Purpose |\n|---|---|\n| **LangChain 0.3** | Prompt templates, chains, LCEL, structured output |\n| **LangGraph** | Stateful agent graphs, multi-node pipelines |\n| **CrewAI** | Role-based multi-agent orchestration |\n| **Google Gemini** | Primary LLM (via `langchain-google-genai`) |\n| **Tavily** | Real-time web search tool for agents |\n| **Chroma** | Local vector store for RAG |\n| **HuggingFace** | Embedding models |\n\n---\n\n## 📈 Learning Progression\n\n```\nLangchain.ipynb ← Start here: prompts + chains\n │\nRag_Implementation.ipynb ← Add knowledge: PDF + vector search\n │\nLangchain_Agents.ipynb ← Add tools: agents + web search\n │\nLanggraph_Agents.ipynb ← Add control flow: graph-based agents\n │\nChatbot_Graph.ipynb ← Add memory: stateful conversations\n │\nPost_Creation_Agent.ipynb ← Add loops: writer-reviewer cycle\n │\nCrew.ipynb ← Add collaboration: multi-agent crews\n │\nCareer_Agent.ipynb ← Final: full career AI agent\n```\n\n---\n\n## 🤝 Contributing\n\nFound a bug or want to add more notebooks? PRs welcome!\n\n---\n\n## 📄 License\n\nMIT © [Ashwin14101](https://github.com/Ashwin14101)\n","readmeExcerpt":"🤖 LangChain & LangGraph — Agents, RAG & Multi-Agent Systems