What it means

You send a request to Bedrock’s API, and Bedrock runs the AI on AWS machines and sends the result back.

You → Bedrock API → (AWS runs & scales servers) → AI model → Response

You don’t do

• Buy GPUs
• Install ML software
• Set up scaling
• Patch or monitor servers

AWS does

• Runs the hardware
• Scales up/down
• Keeps it online
• Secures it

Why it helps

• Start fast
• Less work
• Grows with your app

Example

Pseudo-code:

summary = bedrock.generate("Summarize this:\n" + text)
show(summary)

JavaScript (Node.js):

// npm i @aws-sdk/client-bedrock-runtime
import { BedrockRuntimeClient, InvokeModelCommand } from "@aws-sdk/client-bedrock-runtime";

const br = new BedrockRuntimeClient({ region: "us-east-1" });

const body = {
anthropic_version: "bedrock-2023-05-31",
max_tokens: 60,
messages: [{ role: "user", content: [{ type: "text", text: "Explain Bedrock in one short line." }] }]
};

const res = await br.send(new InvokeModelCommand({
modelId: "anthropic.claude-3-haiku-20240307-v1:0",
contentType: "application/json",
accept: "application/json",