# LangChain Integration Build AI applications with ProgressDB + LangChain. ProgressDB serves as persistent memory for LangChain conversations, enabling chat history and context retention/retrieval purposes. ## Setup 1. Run ProgressDB locally or use hosted instance 2. Create backend endpoint to sign users with backend SDK 3. Use ProgressDB as LangChain memory store 4. Integrate with LangChain chains and agents ## Python Backend Example ```python from progressdb import ProgressDBClient from langchain.memory import ChatMessageHistory from langchain.chat_models import ChatOpenAI from langchain.chains import ConversationChain from langchain.schema import HumanMessage, AIMessage db = ProgressDBClient( base_url="https://api.example.com", api_key="your-api-key" ) class ProgressDBMemory: def __init__(self, thread_id: str, user_id: str): self.thread_id = thread_id self.user_id = user_id def load_memory_variables(self): messages = db.list_thread_messages(self.thread_id, user_id=self.user_id) history = ChatMessageHistory() for msg in messages: if msg.role == "user": history.add_user_message(msg.body["text"]) else: history.add_ai_message(msg.body["text"]) return {"chat_history": history} def save_context(self, inputs, outputs): db.create_thread_message( self.thread_id, {"body": {"text": inputs["input"]}, "role": "user"}, self.user_id ) db.create_thread_message( self.thread_id, {"body": {"text": outputs["response"]}, "role": "assistant"}, self.user_id ) # FastAPI endpoint @app.post("/chat") async def chat(request: ChatRequest): memory = ProgressDBMemory(request.thread_id, request.user_id) llm = ChatOpenAI(temperature=0) chain = ConversationChain(llm=llm, memory=memory) response = chain.run(input=request.message) return {"response": response} ``` ## React Frontend Example ```tsx import React, { useState } from 'react'; import ProgressDBProvider, { useMessages } from '@progressdb/react'; function ChatInterface({ threadId }: { threadId: string }) { const [input, setInput] = useState(''); const [loading, setLoading] = useState(false); const { messages, create } = useMessages(threadId); const sendMessage = async () => { if (!input.trim()) return; setLoading(true); // Save user message await create({ body: { text: input }, role: 'user' }); // Get AI response from LangChain backend const response = await fetch('/api/chat', { method: 'POST', headers: { 'Content-Type': 'application/json' }, body: JSON.stringify({ message: input, threadId, userId: 'user123' }) }); const data = await response.json(); // Save AI response await create({ body: { text: data.response }, role: 'assistant' }); setInput(''); setLoading(false); }; return (

{messages.map(m => (

{m.role}: {m.body.text}

))}

setInput(e.target.value)} placeholder="Type a message..." onKeyPress={(e) => e.key === 'Enter' && sendMessage()} />

); } export default function App() { return ( { const res = await fetch('/api/sign-user', { method: 'POST', body: JSON.stringify({ userId: 'user123' }) }); return res.json(); }} > ); } ``` ## SDKs & their methods available: - [TypeScript SDK](/clients/typescript) - [React SDK](/clients/reactjs) - [Python SDK](/clients/python) - [Node.js SDK](/clients/nodejs)