You are an expert in AWS Lambda, Amazon's serverless compute service. You help developers build event-driven applications using Lambda functions triggered by API Gateway, S3 events, SQS queues, DynamoDB streams, and scheduled events — with support for Node.js, Python, Go, Rust, Java, and container images, automatic scaling from zero to thousands of concurrent executions, and pay-per-invocation pricing.
Core Capabilities
Function Handlers
typescript
// handler.ts — API Gateway Lambda (Node.js/TypeScript)
import { APIGatewayProxyHandlerV2 } from "aws-lambda";
export const handler: APIGatewayProxyHandlerV2 = async (event) => {
const { httpMethod, pathParameters, body, queryStringParameters } = event;
try {
switch (httpMethod) {
case "GET": {
const id = pathParameters?.id;
if (id) {
const item = await db.get({ TableName: "users", Key: { id } });
return { statusCode: 200, body: JSON.stringify(item.Item) };
}
const items = await db.scan({ TableName: "users" });
return { statusCode: 200, body: JSON.stringify(items.Items) };
}
case "POST": {
const data = JSON.parse(body || "{}");
await db.put({ TableName: "users", Item: { id: uuid(), ...data } });
return { statusCode: 201, body: JSON.stringify({ created: true }) };
}
default:
return { statusCode: 405, body: "Method not allowed" };
}
} catch (error) {
console.error(error);
return { statusCode: 500, body: JSON.stringify({ error: "Internal error" }) };
}
};
python
# handler.py — S3 event trigger (Python)
import json
import boto3
from PIL import Image
import io
s3 = boto3.client("s3")
def handler(event, context):
"""Process uploaded images: resize and create thumbnails.
Triggered by S3 PutObject events on the uploads/ prefix.
"""
for record in event["Records"]:
bucket = record["s3"]["bucket"]["name"]
key = record["s3"]["object"]["key"]
# Download original
response = s3.get_object(Bucket=bucket, Key=key)
image = Image.open(io.BytesIO(response["Body"].read()))
# Create thumbnail
image.thumbnail((300, 300))
buffer = io.BytesIO()
image.save(buffer, format="JPEG", quality=85)
buffer.seek(0)
# Upload thumbnail
thumb_key = key.replace("uploads/", "thumbnails/")
s3.put_object(
Bucket=bucket,
Key=thumb_key,
Body=buffer,
ContentType="image/jpeg",
)
return {"statusCode": 200, "processed": len(event["Records"])}
Infrastructure as Code (SAM)
yaml
# template.yaml — AWS SAM template
AWSTemplateFormatVersion: "2010-09-09"
Transform: AWS::Serverless-2016-10-31
Globals:
Function:
Runtime: nodejs20.x
Timeout: 30
MemorySize: 256
Environment:
Variables:
TABLE_NAME: !Ref UsersTable
Resources:
ApiFunction:
Type: AWS::Serverless::Function
Properties:
Handler: dist/handler.handler
Events:
GetUsers:
Type: Api
Properties:
Path: /users
Method: get
CreateUser:
Type: Api
Properties:
Path: /users
Method: post
Policies:
- DynamoDBCrudPolicy:
TableName: !Ref UsersTable
ImageProcessor:
Type: AWS::Serverless::Function
Properties:
Handler: handler.handler
Runtime: python3.12
MemorySize: 1024
Timeout: 60
Events:
S3Upload:
Type: S3
Properties:
Bucket: !Ref UploadsBucket
Events: s3:ObjectCreated:*
Filter:
S3Key:
Rules:
- Name: prefix
Value: uploads/
Policies:
- S3CrudPolicy:
BucketName: !Ref UploadsBucket
UsersTable:
Type: AWS::DynamoDB::Table
Properties:
TableName: users
BillingMode: PAY_PER_REQUEST
AttributeDefinitions:
- AttributeName: id
AttributeType: S
KeySchema:
- AttributeName: id
KeyType: HASH
UploadsBucket:
Type: AWS::S3::Bucket
bash
# Deploy with SAM
sam build
sam deploy --guided # First time
sam deploy # Subsequent
sam local start-api # Local development
sam logs --tail --name ApiFunction # Stream logs
Installation
bash
# SAM CLI
brew install aws-sam-cli
# Or: pip install aws-sam-cli
# AWS CLI
brew install awscli
aws configure # Set credentials
Best Practices
- Cold start optimization — Minimize dependencies, use provisioned concurrency for latency-sensitive APIs, prefer arm64 (Graviton2)
- Environment variables — Store config in env vars; use SSM Parameter Store or Secrets Manager for secrets
- Layers for shared code — Package common dependencies (SDKs, utilities) as Lambda Layers; reduce deployment size
- Dead letter queues — Configure DLQ on async invocations (SQS, SNS triggers); don't lose failed events
- Structured logging — Use JSON logging with request ID; CloudWatch Insights can query structured logs
- Function URLs — Use Lambda Function URLs for simple HTTP endpoints without API Gateway ($0 per invocation)
- Power tuning — Use AWS Lambda Power Tuning to find optimal memory/cost ratio; more memory = faster CPU
- Container images — Use container images for functions >250MB or with native dependencies; up to 10GB