aws-expert
AWS cloud architecture and serverless expertise from Amazon veterans (Distinguished Engineers, Prime Day scale). Use when setting up AWS infrastructure (EC2, ECS, EKS, Lambda, S3, RDS, DynamoDB), desi
SKILL.md.claude/skills/aws-expert/SKILL.md
GitHubAWS Expert Skill
Overview
AWS cloud expertise from Amazon engineers who built and operate the world's largest cloud platform.
What Qualifies as "AWS Expert"
Level 10/10 (Distinguished)
- 12+ years at AWS or using AWS at massive scale
- Designed AWS services or major platform components
- Handles Prime Day level traffic (100M+ req/min)
- Deep knowledge of 50+ AWS services
- Examples: Amazon Distinguished Engineers, AWS Solutions Architects
Level 9/10 (Principal)
- 10+ years of AWS experience
- Architected multi-region, highly available systems
- Expert in cost optimization at scale
- Can design complex microservices on AWS
- Examples: Principal Engineers at Amazon, Netflix, Airbnb
Level 8/10 (Senior)
- 7+ years of AWS experience
- Can architect production-grade systems
- Strong DevOps and automation skills
- Handles millions of users
Key Capabilities
Compute Services
- EC2: Instances, auto-scaling, spot instances
- ECS: Container orchestration
- EKS: Kubernetes on AWS
- Lambda: Serverless functions
- Fargate: Serverless containers
Storage & Database
- S3: Object storage, lifecycle policies, versioning
- RDS: PostgreSQL, MySQL, Aurora
- DynamoDB: NoSQL at scale, GSI/LSI, DAX caching
- ElastiCache: Redis, Memcached
- EBS: Block storage, snapshot strategies
Networking & Content Delivery
- VPC: Networking, subnets, security groups
- CloudFront: CDN, edge computing, Lambda@Edge
- Route53: DNS, health checks, failover
- API Gateway: REST/WebSocket APIs
- ALB/NLB: Load balancing
Developer Tools
- CodePipeline: CI/CD automation
- CodeBuild: Build automation
- CodeDeploy: Deployment strategies
- CloudFormation: Infrastructure as Code
- CDK: AWS Cloud Development Kit
Monitoring & Operations
- CloudWatch: Metrics, logs, alarms, dashboards
- X-Ray: Distributed tracing
- Systems Manager: Operations management
- CloudTrail: Audit logs
Security & Compliance
- IAM: Users, roles, policies
- KMS: Encryption key management
- Secrets Manager: Secret rotation
- WAF: Web application firewall
- Shield: DDoS protection
Notable AWS Architectures at FAANG
Amazon.com
- Prime Day infrastructure (100M+ req/min)
- Multi-region active-active
- Microservices on ECS/EKS
- DynamoDB for cart (scales to zero)
Netflix
- Global streaming on AWS (all workloads)
- 200+ microservices
- Multi-region disaster recovery
- Chaos engineering (Chaos Monkey)
Airbnb
- Payment processing on AWS
- Search on ElasticSearch
- Data pipeline on EMR
Architecture Patterns
High Availability Pattern
┌─────────────────────────────────────┐
│ Route53 (DNS) │
│ Health checks + Failover │
└─────────────┬───────────────────────┘
│
┌─────────────▼───────────────────────┐
│ CloudFront (CDN) │
│ Global edge caching │
└─────────────┬───────────────────────┘
│
┌────▼────┐
│ ALB │
│ (Multi-AZ)
└────┬────┘
┌────────┼────────┐
│ │ │
┌───▼──┐ ┌──▼───┐ ┌─▼────┐
│ ECS │ │ ECS │ │ ECS │
│ (AZ1)│ │(AZ2) │ │(AZ3) │
└──────┘ └──────┘ └──────┘
Serverless Pattern
API Gateway → Lambda → DynamoDB
│
├→ S3 (storage)
└→ SQS (async tasks)
Microservices Pattern
ALB
├→ Service 1 (ECS) → RDS (PostgreSQL)
├→ Service 2 (Lambda) → DynamoDB
└→ Service 3 (ECS) → ElastiCache (Redis)
Cost Optimization Techniques
From Amazon experts:
- Right-sizing: Use Compute Optimizer
- Reserved Instances: 1-year or 3-year for steady workloads
- Spot Instances: 90% discount for fault-tolerant workloads
- Savings Plans: Flexible commitment
- S3 Lifecycle: Move to Glacier, Intelligent-Tiering
- Auto-scaling: Scale down during off-peak
- Lambda: Pay per use, scales to zero
- DynamoDB On-Demand: For unpredictable traffic
Amazon Prime Day example: $2M+ saved through spot instances and pre-warming strategies.
Typical Responsibilities
AWS experts at Amazon/FAANG:
- Design cloud architectures
- Optimize costs (millions in savings)
- Ensure 99.99%+ uptime
- Automate infrastructure (IaC)
- Set up monitoring and alerting
- Security and compliance
- Disaster recovery planning
- Capacity planning
- Performance optimization
When to Need AWS Experts
Projects requiring:
- Cloud infrastructure setup
- High availability (99.99%+)
- Scalability (millions of users)
- Cost optimization
- Microservices on AWS
- Serverless architectures
- Multi-region deployments
- Disaster recovery
- DevOps automation
Related Skills
AWS experts often also have:
- Terraform: Infrastructure as Code
- Docker & Kubernetes: Containers
- Linux: Systems administration
- Networking: VPC, DNS, load balancing
- Security: IAM, encryption, compliance
- Monitoring: CloudWatch, Datadog, Prometheus
- CI/CD: Jenkins, GitHub Actions
- Scripting: Python, Bash