Look something up. Keep moving.

Cloud Computing Fundamentals

A practical mental model for on-demand cloud resources, elastic capacity, measured consumption, and how cloud changes the way infrastructure is acquired and operated.

Benefits of the AWS Cloud

The six classic cloud advantages explained as business and architecture decisions—from variable expense and economies of scale to agility, elasticity, and global reach.

Cloud Service & Deployment Models

IaaS, PaaS, SaaS, cloud, hybrid, and on-premises—organized around the amount of control you keep and the operational responsibility you hand to a provider.

AWS Global Infrastructure

How Regions, Availability Zones, data centers, and edge-oriented infrastructure fit together—and how to use those boundaries for latency and resilience decisions.

AWS Shared Responsibility Model

A precise guide to security 'of' the cloud versus security 'in' the cloud, including how the responsibility boundary changes across different service models.

Scalability & Elasticity

How cloud capacity grows with a workload, how elasticity follows changing demand, and when vertical or horizontal scaling changes the architecture.

High Availability, Fault Tolerance & Resilience

A clear separation of availability, fault tolerance, reliability, and resilience, with the AWS multi-AZ and multi-Region patterns that support them.

AWS Well-Architected Framework

The six Well-Architected pillars and the review mindset AWS uses to evaluate architecture decisions, expose risks, and improve workloads over time.

Cloud Economics & Pay-as-you-go

How cloud consumption changes the cost model, what pay-as-you-go really means, and why elasticity, commitments, and cost discipline determine economic outcomes.

Cloud Adoption, AWS CAF & Migration

A high-level map from business transformation to workload migration: the AWS Cloud Adoption Framework, its six perspectives, four journey phases, and the seven migration strategies.

Amazon EC2

The core AWS virtual-server service: how instances are assembled from images, instance types, networking and storage, what you control, and when EC2 is the right compute abstraction.

EC2 Instance Types & Families

How EC2 families and sizes map workload requirements to CPU, memory, storage, networking and accelerators—and why right-sizing is a measurement problem rather than a naming exercise.

EC2 AMIs, Launch Templates & User Data

Three complementary mechanisms for repeatable EC2 launches: a bootable image, reusable launch parameters, and per-instance bootstrap instructions.

EC2 Purchasing & Capacity Options

A decision guide to On-Demand, Savings Plans, Reserved Instances, Spot, Capacity Reservations and dedicated tenancy—separating cost commitments from capacity guarantees.

Amazon EC2 Auto Scaling

How Auto Scaling groups maintain healthy EC2 capacity, scale between minimum and maximum bounds, and combine launch definitions, metrics and Availability Zones into an elastic fleet.

Elastic Load Balancing: ALB, NLB & GWLB

How AWS load balancers distribute traffic to healthy targets and how Application, Network and Gateway Load Balancers differ by layer, routing behavior and workload purpose.

AWS Lambda

Event-driven serverless compute: invocation, execution environments, concurrency, scaling, permissions, pricing dimensions and the architectural situations where functions fit best.

Containers on AWS: ECS, EKS & Fargate

A clear separation of container orchestration and compute: Amazon ECS, Amazon EKS, EC2 capacity and AWS Fargate, with decision criteria for each layer.

AWS Elastic Beanstalk

An application-platform abstraction that deploys code while provisioning and coordinating underlying AWS resources such as compute, scaling, load balancing and health monitoring.

Amazon Lightsail

A simplified AWS experience for websites, small applications and projects, packaging virtual servers and common supporting resources behind predictable bundles and a focused interface.

AWS Batch

Managed orchestration for batch workloads: job definitions, queues, scheduling and automatically scaled EC2 or Fargate compute for finite units of work.

AWS Outposts

AWS-managed infrastructure deployed at a customer site, extending selected AWS services and APIs on premises for low-latency, local-processing and hybrid requirements.

AWS Storage Models: Object, Block & File

A decision-first comparison of object, block and file storage: how applications address data, what access semantics they expect, and which AWS services map to each model.

Amazon S3

AWS object storage from first principles: buckets, objects, keys, regional placement, consistency, access patterns and the architectural situations in which S3 is the natural storage choice.

S3 Storage Classes & S3 Glacier

How S3 storage classes trade storage cost, access pattern, resilience and retrieval behavior—from frequently accessed objects to long-term archival data.

S3 Versioning, Lifecycle & Replication

Three complementary S3 mechanisms for recovering old object states, automating retention and storage-class movement, and maintaining additional copies across buckets or Regions.

S3 Security & Access Control

A practical S3 security model covering identities, bucket policies, Object Ownership, Block Public Access, encryption and why public access must be an explicit architecture decision.

Amazon EBS

Durable block storage for EC2: Availability-Zone placement, attachment semantics, persistence, encryption, snapshots and how EBS differs from instance store and shared storage.

EBS Volume Types & Snapshot Strategy

How General Purpose SSD, Provisioned IOPS SSD and HDD-backed EBS options map to random I/O, throughput and cost—and how snapshots change the recovery model.

Amazon EFS

Managed elastic NFS file storage for AWS compute: shared file-system semantics, mount targets, Regional versus One Zone storage and the workloads EFS is designed to simplify.

Amazon FSx

Managed feature-rich file systems on AWS: how FSx for Windows File Server, Lustre, NetApp ONTAP and OpenZFS map existing file-system requirements to managed cloud storage.

AWS Storage Gateway

Hybrid storage that connects on-premises applications to AWS storage through file, volume and virtual-tape interfaces while retaining local caching where the gateway model requires it.

AWS Backup

Centralized policy-based data protection across supported AWS services: backup plans, vaults, retention, copy jobs, restore workflows and the distinction between backup and availability.

AWS Elastic Disaster Recovery

Server-level disaster recovery through continuous block replication, a low-cost staging area, recovery launches and point-in-time recovery for on-premises or cloud-based source servers.

Amazon VPC, CIDR & Subnets

The network boundary underneath AWS workloads: VPC address space, CIDR planning, subnets, Availability Zones, IPv4/IPv6 and the distinction between network placement and internet reachability.

VPC Routing, Internet Gateway & NAT Gateway

How packets leave a subnet: destination-to-target routes, longest-prefix matching, internet gateways, NAT gateways and the architecture of public, private and isolated IPv4 tiers.

Security Groups vs. Network ACLs

Two distinct VPC traffic controls: stateful resource-level security groups and stateless subnet-level network ACLs, including direction, rule behavior and practical troubleshooting.

VPC Peering & AWS Transit Gateway

Private network-to-network connectivity from direct VPC peering to hub-and-spoke Transit Gateway architectures, with CIDR constraints, route propagation and scale trade-offs.

VPC Endpoints & AWS PrivateLink

Private paths from a VPC to AWS and endpoint services: gateway endpoints, interface endpoints, PrivateLink mechanics, DNS behavior and when a NAT or internet gateway is unnecessary.

Amazon Route 53 & DNS Routing

AWS DNS from hosted zones and records to alias targets, health checks and routing policies for latency, failover, weighted and location-aware traffic decisions.

Amazon CloudFront & Edge Caching

AWS content delivery at the edge: distributions, origins, cache behavior, cache keys, TTL controls, cache hit/miss flow and how CloudFront differs from DNS and Global Accelerator.

Dedicated private connectivity from a customer network to AWS: Direct Connect locations, virtual interfaces, BGP routing, Direct Connect gateways and the difference between private connectivity and encryption.

AWS VPN: Site-to-Site & Client VPN

Encrypted hybrid and remote-user connectivity: Site-to-Site IPsec tunnels, customer and AWS gateways, dynamic BGP routing and managed Client VPN remote access.

AWS Global Accelerator

Global anycast entry points for TCP/UDP applications: static IP addresses, listeners, endpoint groups, health-based routing, traffic dials and the AWS global network.

Amazon API Gateway

Managed front doors for REST, HTTP and WebSocket APIs: routes and methods, stages, integrations, authorization, request throttling and the feature trade-off between REST and HTTP APIs.

Choosing an AWS Database & Data Model

A workload-first framework for choosing among relational, key-value, document, in-memory, and analytical data stores on AWS by examining access patterns, transactions, consistency, scale, latency, and operational requirements.

Amazon RDS

The managed relational database service for familiar database engines, with a clear division between database-level choices you still own and infrastructure operations such as provisioning, backups, patching, monitoring integration, and high-availability options that RDS manages.

Amazon Aurora

AWS's MySQL- and PostgreSQL-compatible relational database architecture, separating compute instances from a distributed cluster volume and using Aurora Replicas, cluster endpoints, and serverless capacity options to address availability and scaling.

RDS Multi-AZ vs. Read Replicas

A precise comparison of RDS availability and read-scaling mechanisms: synchronous or semisynchronous Multi-AZ deployment patterns, readable Multi-AZ DB cluster standbys, asynchronous read replicas, failover, replica lag, and promotion.

Amazon DynamoDB Fundamentals

The core DynamoDB mental model: tables, items, partition and sort keys, access-pattern-driven design, Query versus Scan behavior, physical partition distribution, and why key design determines both performance and scalability.

DynamoDB Capacity, Consistency & Indexes

The DynamoDB mechanics behind throughput and alternate access patterns: on-demand versus provisioned capacity, read/write unit calculations, eventual and strong reads, global and local secondary indexes, and the capacity consequences of each choice.

Amazon ElastiCache

A managed in-memory data store and cache layer using Valkey, Redis OSS, or Memcached, with clear patterns for cache-aside acceleration, session/state workloads, replication, sharding, eviction, and the operational consequences of treating memory as a performance tier.

Amazon DocumentDB

A managed document database with MongoDB compatibility, built around clusters with a primary, read replicas, distributed cluster storage, document indexes, TTL behavior, and explicit compatibility considerations for MongoDB applications.

Amazon Redshift & the Analytics Boundary

How Amazon Redshift differs from an operational relational database: columnar storage, massively parallel query execution, distribution and sort choices, compression, managed storage, and the workloads that belong in a data warehouse rather than an OLTP path.

AWS Database Migration Service (AWS DMS)

The mechanics of moving databases with minimal downtime: endpoints/data providers, full load, cached changes, change data capture, validation, homogeneous versus heterogeneous migrations, and the separate job of schema conversion.

AWS IAM Authorization & Least Privilege

How AWS decides whether a request is allowed: principals, identity and resource policies, explicit deny, permissions boundaries, organization guardrails, conditions and a practical route toward least privilege.

IAM Roles, AWS STS & Temporary Credentials

The preferred AWS pattern for delegated and workload access: role trust, AssumeRole, short-lived sessions, session policies, role chaining and traceable temporary credentials instead of embedded long-term keys.

IAM Identity Center & Workforce Federation

Central workforce access to AWS accounts and applications: identity sources, SAML/SCIM federation, permission sets and the IAM roles that turn a workforce assignment into temporary AWS account access.

AWS KMS, Envelope Encryption & Key Access

AWS key management from encryption keys and envelope encryption to key policies, grants, encryption context and the cryptographic choices that matter when services protect data with KMS.

AWS Secrets Manager

Secure storage and controlled retrieval of application secrets, including version staging labels, KMS encryption, rotation workflows, resource policies and the design reason to remove credentials from code.

AWS Certificate Manager & TLS Certificates

Managed X.509 certificate lifecycle for HTTPS and TLS: public versus private certificates, domain validation, regional placement, renewal and the distinction between certificate management and web-request filtering.

AWS WAF & AWS Shield

Application-layer request filtering versus managed DDoS protection, with web ACL evaluation order, terminating actions, managed rule groups, rate-based controls, WCU capacity and Shield Standard versus Shield Advanced.

Amazon GuardDuty

Managed threat detection for AWS accounts and workloads: foundational telemetry, protection-plan data sources, threat intelligence, machine learning, findings and the difference between detecting a threat and enforcing preventive controls.

Amazon Inspector

Continuous vulnerability management for supported AWS compute workloads, including EC2, ECR container images and Lambda, with CVE-aware findings, reachability context and severity scoring.

Amazon Macie

Sensitive-data discovery for Amazon S3: automated discovery, targeted jobs, managed and custom data identifiers, regex and proximity criteria, findings, allow lists and protected result storage.

Security Audit Logging: CloudTrail, Config & CloudWatch

Three evidence planes that are often confused: CloudTrail for API activity, AWS Config for resource configuration history and compliance, and CloudWatch for operational metrics, logs and alarms—plus cryptographic CloudTrail log validation.

AWS Organizations SCP Security Guardrails

Preventive authorization guardrails across AWS accounts: organization hierarchy, service control policy inheritance, allow-list versus deny-list models, root and service-linked-role behavior, and why SCPs set maxima rather than grant permissions.

Amazon CloudWatch: Metrics, Logs & Alarms

AWS operational observability through metrics, dimensions, log groups, log streams and alarms—including the evaluation settings that turn telemetry into actionable state changes.

AWS Organizations, OUs & Service Control Policies

Multi-account governance through an organization root, organizational units, member accounts and SCP permission guardrails, with delegated administration and centralized policy inheritance.

AWS Control Tower & Landing Zones

A prescriptive multi-account landing zone that orchestrates AWS Organizations and other governance services, standardizes account provisioning, and applies preventive, detective and proactive controls.

AWS Health: Account & Service Events

Personalized operational awareness for AWS service issues and planned changes, with affected-resource context, organization aggregation and event-driven automation.

AWS Trusted Advisor

Automated AWS best-practice checks and recommendations across cost, performance, security, fault tolerance, service limits and operational excellence, with plan-dependent access and organization views.

AWS Well-Architected Tool & Operational Governance

A repeatable architecture-review process using workloads, the six Well-Architected pillars, lenses, risk findings, improvement plans and milestones to govern architecture over time.

AWS CloudTrail: Event History & Trails

The AWS account audit record: management, data, network activity and Insights events, the built-in Event history, and durable multi-Region or organization trails.

AWS Config: Resource History & Compliance

Configuration-state governance for AWS resources: recorders and configuration items, Config rules, conformance packs, remediation and multi-account aggregators.

AWS Systems Manager: Fleet Operations

Central operations for managed nodes and resources, with Session Manager, Run Command, State Manager, Patch Manager, Maintenance Windows and Parameter Store.

AWS CloudFormation & Infrastructure as Code

Declarative AWS infrastructure through templates and stacks, with parameters, change sets, update behavior, drift detection and StackSets for repeated multi-account deployment.

AI, Machine Learning & Generative AI Fundamentals

A durable map of artificial intelligence, machine learning, deep learning, generative AI, and agentic AI—what each term means, where the boundaries overlap, and when prediction is a better fit than generation.

Supervised, Unsupervised & Reinforcement Learning

The three core machine-learning paradigms organized around the feedback available to the learner: labels, latent structure, or rewards from interaction with an environment.

Machine Learning Lifecycle & MLOps

From framing a measurable problem through data preparation, training, evaluation, deployment, monitoring, feedback, and retraining—with the controls needed to make ML repeatable in production.

Amazon SageMaker AI

AWS's managed machine-learning platform for preparing data, training models at scale, tracking and governing ML assets, and deploying inference—best understood as lifecycle building blocks rather than one monolithic model service.

Foundation Models, Tokens, Embeddings & Context

The technical vocabulary behind modern generative AI: foundation models, tokenization, context windows, embeddings, vector similarity, multimodality, and the difference between generating content and representing meaning.

Amazon Bedrock

The fully managed AWS service for building generative-AI applications with foundation models through managed inference APIs plus capabilities for retrieval, guardrails, evaluation, customization, and agentic systems.

Prompt Engineering & Inference Parameters

How instructions, context, examples, templates, and sampling controls shape foundation-model behavior—and why temperature, top-p, top-k, maximum tokens, and stop sequences must be interpreted per model.

RAG, Vector Retrieval & Bedrock Knowledge Bases

How Retrieval Augmented Generation grounds foundation-model answers in external data using chunking, embeddings, vector or hybrid retrieval, prompt augmentation, and source-aware evaluation.

AI Agents, Tool Use & Amazon Bedrock AgentCore

How agentic systems turn model reasoning into controlled actions using tools, orchestration, memory, identity, and observations—with current AWS direction centered on Bedrock AgentCore.

Foundation Model Customization & Fine-Tuning

A decision framework for in-context learning, RAG, supervised fine-tuning, continued pretraining, reinforcement-based tuning, and model distillation—what changes model weights and what does not.

Model Evaluation & AI Metrics

How to evaluate predictive ML, foundation models, RAG, and agents with task-appropriate metrics, representative datasets, human judgment, LLM-as-a-judge, and business outcome measures.

Responsible AI, Security, Guardrails & Governance

A layered approach to trustworthy AI that separates fairness, explainability, robustness, privacy, and governance from hard security controls such as IAM, encryption, logging, data boundaries, and Bedrock Guardrails.

Advanced RAG Engineering

Production retrieval-augmented generation beyond the basic pattern: chunking, hybrid retrieval, metadata filters, query decomposition, reranking, citations, and stage-by-stage quality tuning.

Vector Stores for Generative AI on AWS

How embeddings become searchable infrastructure: similarity metrics, HNSW-style indexes, metadata, Aurora PostgreSQL with pgvector, managed retrieval choices, scaling, and index maintenance.

Amazon Bedrock Production Inference

Serving foundation models under real traffic: Converse and InvokeModel, streaming, inference profiles, Cross-Region Inference, Provisioned Throughput, prompt caching, quotas, latency, and graceful degradation.

AgentCore, Strands Agents, MCP & Multi-Agent Systems

Production agentic architecture on AWS: reasoning loops, tool contracts, Model Context Protocol, AgentCore Gateway, Strands Agents, multi-agent coordination, identity, stopping conditions, and tracing.

Bedrock Prompt Management & Flows

Treat prompts as versioned application assets: variables, variants, inference configuration, reusable versions, flow nodes, conditional branches, iteration, testing, and regression-safe promotion.

GenAI API & Streaming Architectures

Design the application boundary around long-running and token-streaming inference: synchronous versus asynchronous work, SSE/WebSockets, queues, backpressure, API throttling, retries, and request correlation.

GenAI CI/CD & Model Lifecycle

Release generative AI safely by versioning prompts, models, adapters, retrieval configuration, evaluation evidence, and infrastructure together—with quality gates, canaries, rollback, and retirement.

Multimodal GenAI Data Pipelines

Prepare documents, images, audio, video, and structured data for generative applications while preserving provenance, confidence, metadata, privacy, and modality-specific information.

Prompt Injection, Jailbreaks & GenAI Threat Modeling

Threat-model generative and agentic systems around untrusted instructions, prompt leakage, excessive agency, unsafe tools, poisoned context, Guardrails, validation, least privilege, and adversarial testing.

GenAI Privacy & Data Protection

Protect prompts, retrieved context, model outputs, embeddings, logs, and evaluation datasets with classification, minimization, PII detection, masking, encryption, private connectivity, retention, and tenant-aware authorization.

AI Governance, Lineage & Model Cards

Make GenAI decisions auditable through model cards, intended-use boundaries, dataset and prompt lineage, source attribution, change records, approval evidence, CloudTrail, and continuous governance checks.

GenAI Cost & Token Optimization

Optimize spend without sacrificing quality: token accounting, context pruning, prompt compression, response limits, model routing, prompt/semantic caching, batching, concurrency, and capacity-mode decisions.

GenAI Observability

Observe the complete GenAI request path across tokens, latency, errors, model identity, retrieval, safety, agents, tools, cost, and quality—with Bedrock invocation logs, CloudWatch, traces, and actionable SLOs.

Advanced GenAI Evaluation

Build repeatable evidence for model, RAG, and agent quality with golden datasets, LLM-as-a-judge, human review, retrieval metrics, task completion, A/B and canary comparisons, and release thresholds.

GenAI Testing & Troubleshooting

A systematic playbook for context overflow, truncation, API failures, prompt regressions, embedding drift, poor retrieval, malformed structured output, agent/tool failures, and production quality regressions.

AWS Analytics Service Selection

A decision framework for choosing among streaming ingestion, catalog and ETL, serverless SQL, open-source big-data processing, data warehousing, and business intelligence on AWS.

Amazon Kinesis Data Streams

Real-time record streaming with partition keys, shards, sequence numbers, consumer models, retention, and resharding as the core scaling concepts.

AWS Glue: Data Catalog, Crawlers & ETL

Serverless data integration with a persistent metadata catalog, crawlers and classifiers for discovery, and managed ETL jobs for transforming analytical datasets.

Amazon Athena: Serverless SQL Analytics

Interactive SQL over data in Amazon S3 and supported federated sources, with Glue Catalog metadata, partitions, columnar formats, workgroups, and scan reduction as the key design levers.

Amazon EMR: Big-Data Processing

Managed execution for Spark, Hadoop, and related open-source analytics frameworks, with clusters, Serverless and EKS deployment choices plus storage and scaling considerations.

Amazon Redshift: Analytical Data Warehousing

Managed analytical SQL with columnar storage, distributed execution, distribution and sort strategies, Serverless, and Spectrum for querying S3 beyond the warehouse boundary.

Amazon Quick Sight: Business Intelligence & Dashboards

AWS business intelligence within Amazon Quick, covering datasets, analyses, visuals, dashboards, direct-query access, and the SPICE in-memory engine, with QuickSight still used in API and certification references.

Amazon SQS: Queues, Visibility & DLQs

Durable asynchronous queues with Standard versus FIFO semantics, visibility timeout, long polling, dead-letter queues, message groups, and idempotent consumer design.

Amazon SNS: Pub/Sub & Fanout

Push-based topic messaging for one-to-many fanout, subscription filtering, SQS buffering, and Standard versus FIFO topic semantics.

Amazon EventBridge: Event Routing

Event-driven routing through event buses, rules, JSON event patterns and targets, with Pipes for point-to-point integrations and cross-account event architectures.

AWS Step Functions: Workflow Orchestration

State-machine orchestration with Amazon States Language, Task and control-flow states, Standard versus Express execution, service integrations, retries, catches, timeouts, and Map concurrency.

SQS vs. SNS vs. EventBridge vs. Step Functions

A practical boundary between queue buffering, topic fanout, content-based event routing, and stateful orchestration, including the patterns where these services work best together.

AWS Pricing & Billing Fundamentals

A practical model for reading an AWS bill: usage dimensions, variable consumption, tiered rates, data transfer, commitments, and the difference between price, cost, and value.

AWS Cost Explorer

Interactive cost and usage analysis with filters, groups, historical trends, forecasts, commitment reports, and multiple cost metrics.

AWS Budgets: Cost, Usage, Coverage & Utilization

Financial guardrails that compare actual or forecasted AWS consumption with defined thresholds and can notify or trigger supported budget actions.

AWS Data Exports & CUR 2.0

The detailed cost-data layer for recurring analysis: CUR 2.0, consistent schemas, selected columns, row filters, S3 delivery, and downstream financial analytics.

AWS Pricing Calculator & Cost Estimation

Pre-deployment cost modelling with explicit workload assumptions, service configurations, Regions, pricing options, and architecture groups.

Cost Allocation: Tags, Accounts & Cost Categories

How AWS financial dimensions turn raw spend into ownership: linked accounts, activated cost-allocation tags, Cost Categories, showback, chargeback, and shared-cost allocation.

AWS Organizations & Consolidated Billing

One multi-account billing family with centralized payment, combined usage, linked-account visibility, and sharing of eligible volume and commitment discounts.

AWS Compute Optimizer & Rightsizing

Metric-driven recommendations for reducing idle or oversized resources while preserving workload performance and capacity requirements.

Savings Plans, Reservations & Spot Economics

How AWS discount models trade flexibility, commitment, and interruption tolerance—and how utilization and coverage reveal whether the portfolio is efficient.

AWS Support Plans & Trusted Advisor

The financial and operational boundary between included self-service guidance, paid technical support, Trusted Advisor recommendations, AWS Health, and proactive expert engagement.

Cloud Financial Management & Unit Economics

A FinOps-style operating model for connecting AWS spend to business outcomes through ownership, budgets, forecasts, optimization, and cost per unit of value.