add node gen ai sample app

Author: liustve

.github/workflows/node-sample-app-s3-deploy.yml

@@ -3,6 +3,9 @@
 
 name: Sample App Deployment - Node S3
 on:
+  push:
+    branches:
+      - add-genai-sample-apps
   workflow_dispatch: # be able to run the workflow on demand
 
 permissions:

sample-apps/node/genai-service/customInstrumentation.js

@@ -0,0 +1,28 @@
+const { trace } = require('@opentelemetry/api');
+const { registerInstrumentations } = require('@opentelemetry/instrumentation');
+const { HttpInstrumentation } = require('@opentelemetry/instrumentation-http');
+const tracerProvider = trace.getTracerProvider();
+const {
+  LangChainInstrumentation,
+} = require("@traceloop/instrumentation-langchain");
+
+const AgentsModule = require('langchain/agents');
+const ChainsModule = require('langchain/chains');
+const RunnableModule = require('@langchain/core/runnables')
+const ToolsModule = require('@langchain/core/tools')
+const VectorStoresModule = require('@langchain/core/vectorstores')
+
+const traceloop = require("@traceloop/node-server-sdk")
+
+traceloop.initialize({
+  appName: "myTestApp",
+  instrumentModules: {
+    langchain: {
+      runnablesModule: RunnableModule,
+      toolsModule: ToolsModule,
+      chainsModule: ChainsModule,
+      agentsModule: AgentsModule,
+      vectorStoreModule: VectorStoresModule,
+    }
+  }
+});

sample-apps/node/genai-service/index.js

@@ -0,0 +1,64 @@
+const express = require('express');
+const { BedrockChat } = require("@langchain/community/chat_models/bedrock");
+const { ChatPromptTemplate } = require("@langchain/core/prompts");
+const traceloop = require("@traceloop/node-server-sdk")
+const logger = require('pino')()
+
+const app = express();
+app.use(express.json());
+const PORT = parseInt(process.env.SAMPLE_APP_PORT || '8000', 10);
+
+const llm = new BedrockChat({
+    model: "anthropic.claude-3-sonnet-20240229-v1:0",
+    region: "us-east-1",
+    credentials: {
+        accessKeyId: process.env.BEDROCK_AWS_ACCESS_KEY_ID,
+        secretAccessKey: process.env.BEDROCK_AWS_SECRET_ACCESS_KEY,
+    },
+    temperature: 0.7,
+});
+
+const prompt = ChatPromptTemplate.fromMessages([
+    [
+      "system",
+      "You are a helpful assistant. Provide a helpful response to the following user input.",
+    ],
+    ["human", "{input}"],
+  ]);
+
+const chain = prompt.pipe(llm);
+
+app.get('/health', (req, res) => {
+    res.json({ status: 'healthy' });
+});
+
+app.post('/ai-chat', async (req, res) => {
+    const { message } = req.body;
+    
+    if (!message) {
+        return res.status(400).json({ error: 'Message is required' });
+    }
+
+    try {
+        logger.info(`Question asked: ${message}`);
+        
+        const response = await traceloop.withWorkflow({ name: "sample_chat" }, () => {
+            return traceloop.withTask({ name: "parent_task" }, () => {
+                return chain.invoke({
+                    input_language: "English",
+                    output_language: "English",
+                    input: message,
+                });
+            });
+        });
+        
+        res.json({ response: response.content });
+    } catch (error) {
+        logger.error(`Error processing request: ${error.message}`);
+        res.status(500).json({ error: 'Internal server error' });
+    }
+});
+
+app.listen(PORT, () => {
+    logger.info(`GenAI service listening on port ${PORT}`);
+});