AWS CloudOps Engineer

You cannot operate what you cannot measure. You collect system and application metrics, set alarms that fire on real problems, and build dashboards that a team can read at a glance.

• CloudWatch metrics and namespaces
• The CloudWatch agent for memory and disk
• Static and anomaly-detection alarms
• Composite alarms and alarm actions
• EC2 status checks and instance recovery
• Dashboards for operators

You build A dashboard and alarm set for a web tier and its database.

Logs explain what metrics only hint at. You collect and search logs, turn patterns into metrics, and route events and notifications so the right person hears about the right problem.

• CloudWatch Logs groups, retention and Logs Insights
• Metric filters that count log patterns
• Subscription filters and log delivery to storage
• CloudTrail for who did what
• EventBridge rules and SNS notifications
• Cross-account log collection

You build A logging setup that turns three log patterns into alarms and notifications.

Knowing something is wrong is half the job. You read health and availability signals, decide the cause from the evidence, and apply fixes by hand or through automation.

• AWS Health events and service status
• Reading load balancer and instance availability metrics
• Automatic instance recovery and replacement
• Systems Manager Automation for common fixes
• Runbooks and escalation
• Verifying a fix worked

You build A runbook and automation that remediates a failed web instance.

Slow systems cost money twice, in resources and in users. You size instances, choose storage types, inspect database performance and add caching where it pays.

• Right-sizing instances with Compute Optimizer
• EBS volume types, IOPS and throughput
• S3 request patterns and transfer speed
• RDS Performance Insights and parameter groups
• Caching with ElastiCache and CloudFront
• Finding the bottleneck before changing anything

You build A performance review of one application with three measured improvements.

Load changes faster than people can react. You scale compute, containers and data stores automatically, choose scaling signals that match demand, and avoid scaling loops that waste money.

• Auto Scaling groups: launch templates and policies
• Target tracking, step and scheduled scaling
• Load balancer health checks and draining
• Scaling DynamoDB and Aurora
• Container and serverless scaling basics
• Testing scaling behavior

You build An Auto Scaling setup that grows and shrinks under a load test.

A system is resilient when a failure is routine. You spread resources across zones, configure failover for compute, databases and DNS, and check that each layer fails over as intended.

• Multi-AZ design for compute and databases
• Load balancers and Route 53 health checks
• Failover routing policies
• Multi-Region basics for critical systems
• Removing single points of failure
• Testing failover

You build A multi-AZ environment with a tested failover for web and database tiers.

A backup you have never restored is a hope. You schedule and protect backups of every data type, copy them across accounts and Regions, and practice restores against stated recovery targets.

• Snapshots, AMIs and AWS Backup plans
• Point-in-time recovery for databases
• S3 versioning, replication and lifecycle
• Cross-Region and cross-account copies
• Recovery objectives and disaster recovery patterns
• Restore testing and reporting

You build A backup plan across three resource types with a documented restore test.

Manual changes drift and cannot be repeated. You provision with CloudFormation, manage fleets with Systems Manager, and automate patching, configuration and routine maintenance.

• CloudFormation stacks, parameters and drift detection
• StackSets and Service Catalog
• Patch Manager and maintenance windows
• State Manager and configuration compliance
• Scheduled and event-driven automation
• Image management with AMIs

You build A CloudFormation environment with automated patching and drift checks.

Operators apply security every day, not only architects. You enforce access and encryption, use detection and compliance tools, and keep fleets patched and configured within policy.

• IAM roles, policies and credential hygiene
• Encryption with KMS for storage and databases
• AWS Config rules and compliance reporting
• GuardDuty, Inspector and Security Hub findings
• Trusted Advisor checks
• Protecting infrastructure: security groups and patching

You build A compliance setup that detects and reports three kinds of misconfiguration.

Many outages are network faults in disguise. You configure VPC connectivity, DNS records and content delivery, then trace connectivity problems from the client to the resource.

• VPC routing, gateways and endpoints
• Route 53 records and routing policies
• CloudFront distributions and cache invalidation
• VPN and Direct Connect connectivity checks
• Flow Logs and Reachability Analyzer
• A method for tracing a connectivity fault

You build A connectivity troubleshooting guide with three traced and fixed faults.

Run one workload as an operations team would. You deploy it, add monitoring, scaling and backups, secure it, then fix a series of faults without taking it down.

• Deploying the workload as code
• Adding monitoring, scaling and backups
• Applying security and compliance checks
• Responding to injected faults
• Writing the operations handbook

You build A running workload with alarms, scaling, backups and an operations handbook.