AWS with Python preview

Use boto3 (AWS SDK for Python) from Python apps to call AWS APIs—S3 uploads, DynamoDB queries, SQS messages. Prefer IAM roles over static keys on servers.

Install and configure

pip install boto3
# Credentials from environment, ~/.aws/credentials, or IAM role
export AWS_PROFILE=sandbox
export AWS_DEFAULT_REGION=us-east-1

List S3 buckets with boto3

import boto3

s3 = boto3.client('s3')
for bucket in s3.list_buckets().get('Buckets', []):
    print(bucket['Name'])

Practice: Run SDK examples locally with sandbox credentials via AWS_PROFILE=sandbox. Never commit real keys—use IAM roles in deployed environments.

Best practices

• Reuse clients/resources at module level (connection pooling)
• Use resource for higher-level S3/DynamoDB; client for full API
• Handle ClientError with explicit error codes
• On Lambda/EC2, attach IAM role—no keys in environment files in Git

Django apps often use boto3 for S3 media storage via django-storages.

Upload object example

s3.upload_file(
    'hello.txt',
    'my-unique-learning-bucket-12345',
    'uploads/hello.txt',
    ExtraArgs={'ServerSideEncryption': 'AES256'},
)

Important interview questions and answers

1. Q: boto3 client vs resource?
   A: Client is low-level API wrapper; resource provides object-oriented interface for select services.
2. Q: Where credentials come from on EC2?
   A: Instance profile IAM role via instance metadata—temporary rotating credentials.

Self-check

1. How should a Django app on EC2 authenticate to S3?
2. What pip package is the AWS SDK for Python?

Pitfall: Creating boto3 client per HTTP request—reuse at module scope.