Skip to content

Latest commit

 

History

History
245 lines (196 loc) · 12.5 KB

index.md

File metadata and controls

245 lines (196 loc) · 12.5 KB

Modern Cloud-Native Serverless Framework on AWS

What Is This Framework?

A comprehensive, opinionated framework for building enterprise-grade cloud-native applications on AWS. It covers everything from architecture principles and software design patterns to CI/CD pipelines, observability, AI integration, and hands-on workshops.

This is not just documentation — it's a decision framework that tells you:

  • What to build with (services, tools, patterns)
  • How to structure code (hexagonal architecture, DDD)
  • How to deploy safely (TPF, canary, Express mode)
  • How to govern at scale (SCPs, RCPs, policy-as-code)
  • How to operate (frontier agents, observability, drift detection)
  • How to develop with AI (AI-DLC, Kiro, ThothCTL)

Framework Components

flowchart TD
    subgraph Principles["PRINCIPLES & PATTERNS"]
        ARCH["Architecture Principles<br/>(16-Factor, Well-Architected, Cell-Based)"]
        SW["Software Architecture<br/>(Hexagonal, DDD, CQRS, Vertical Slice)"]
    end
    subgraph Infrastructure["INFRASTRUCTURE"]
        COMPUTE["Compute<br/>(Lambda, ECS, Step Functions)"]
        IAC["IaC<br/>(CDK + Express, SAM, SST)"]
        API["API & Events<br/>(AppSync, EventBridge, VPC Lattice)"]
        DATA["Data Layer<br/>(DynamoDB, Aurora, ElastiCache)"]
        FRONT["Frontend<br/>(Amplify, SST, Edge)"]
    end
    subgraph Delivery["DELIVERY & OPERATIONS"]
        CICD["CI/CD<br/>(TPF, Ten Pillars, CDK Pipelines)"]
        OBS["Observability<br/>(OpenTelemetry, ADOT, Powertools)"]
        SEC["Security<br/>(Continuum, Guardrails, RCPs)"]
        AIOPS["AI Operations<br/>(DevOps Agent, FinOps Agent)"]
    end
    subgraph Intelligence["INTELLIGENCE LAYER"]
        BRAIN["Custom AI Brain<br/>(Knowledge Bases, AgentCore Memory, MCP)"]
        AGENTS["Agentic Platform<br/>(Strands SDK, AgentCore Runtime)"]
    end
    subgraph Methodology["METHODOLOGY & TOOLS"]
        AIDLC["AI-DLC<br/>(Inception → Construction → Operations)"]
        THOTH["ThothCTL<br/>(Platform Engineering CLI)"]
        KIRO["Kiro<br/>(AI-assisted development)"]
    end
    subgraph Workshops["HANDS-ON WORKSHOPS"]
        W1["Serverless<br/>(Lambda + CDK)"]
        W2["Enterprise<br/>(Constructs + CodeArtifact)"]
        W3["ECS Backend<br/>(Containers + Express Mode)"]
        W4["CI/CD<br/>(Pipelines + Agents)"]
    end

    Principles --> Infrastructure
    Infrastructure --> Delivery
    Delivery --> Intelligence
    Intelligence --> Methodology
    Methodology --> Workshops
Loading

Quick Start

# Install the platform CLI
pip install thothctl

# Initialize development environment (installs all tools)
thothctl init env

# Scaffold a new project
git clone https://github.com/thothforge/cdkv2_typescript_scaffold.git my-project
cd my-project && npm install

# Start AI-DLC workflow with Kiro
kiro-cli chat
> Using AI-DLC, build a serverless order processing API...

Documentation Index

This index follows the same six-group learning path as the site navigation (defined in zensical.toml). Read top-to-bottom for a guided path, or jump to the group you need.

Principles & Patterns

Document Description
Architecture Principles 16-Factor App, Well-Architected, TPF, Platform Engineering, Cell-Based
Software Architecture Hexagonal, Vertical Slice, CQRS, Event Sourcing, DDD — when to use each
Platform Engineering Manifesto, golden paths, self-service, DX metrics, maturity model
IDP → ADP Four levels of agentic maturity, 3 path types, 8 core paths
AI-SDLC AWS AI-DLC methodology, Agentic SDLC, 5-layer reference architecture
DevOps & CI/CD (TPF) Ten Pillars of Pragmatic Deployments, DORA metrics, canary, Express mode
Testing Strategy Testing pyramid, hexagonal testing, contract, chaos, load
APIOps Model Contract-first, API lifecycle, OpenAPI/AsyncAPI/GraphQL, governance
Data Strategy Cell data ownership, ECST, data mesh, data contracts, DynamoDB patterns

Infrastructure

Document Description
Compute Layer Lambda (SnapStart, ARM64, streaming), Step Functions, ECS Fargate
Infrastructure as Code CDK + Express Mode, SAM, SST v3, Terraform — post-Express comparison
CloudFormation Express How Express mode changes the IaC landscape
API & Event Patterns API Gateway, AppSync, EventBridge (Bus, Pipes, Scheduler), VPC Lattice
Data Layer DynamoDB, Aurora Serverless v2, ElastiCache, OpenSearch, S3 Express
Frontend & Full-Stack Amplify Gen 2, SST v3, CloudFront Functions, Lambda@Edge
AI/ML Integration Bedrock, AgentCore, Strands SDK, Guardrails, RAG patterns

Operations & Governance

Document Description
Observability OpenTelemetry (ADOT), Lambda Powertools, Application Signals, Frontier Agents
Security Baseline Defense-in-depth, IAM, zero-trust, data classification, AI security, secrets
Multi-Account Landing Zone Control Tower, SCPs, RCPs, permission boundaries, networking, centralized logging
FinOps & Cost Governance Tagging, service-level optimization, FinOps Agent, cost gates
Disaster Recovery & Continuity RTO/RPO tiers, serverless DR patterns, failover runbooks
Ephemeral Environments & Concurrency Per-developer sandboxes, per-PR environments, agent isolation, TTL auto-cleanup, cost guardrails
Reference Architecture E-commerce, SaaS, IoT — complete stack diagrams
Recommendations Final recommendations by team type and application

Adoption & Onboarding

Document Description
Adoption Roadmap Why this framework, framework comparison, staged roadmap, checklists
Developer Onboarding Prerequisites, first-30-minutes, architecture, dev workflow

Intelligence & AI Brains

Document Description
Custom Intelligent AI Brains How organizations build shared, learning memory layers (AI Brains) to support the entire product lifecycle — architecture, implementation on AWS, case studies (Peloton, Meta, McKinsey), and adoption roadmap
Team Topologies, Roles & Permissions Modern team structures, new AI-era roles (Intent Engineer, AgentOps, Context Engineer), 3-tier decision authority, RACI matrix, AWS permissions model (IAM + Cedar), and staged adoption path

Workshops

Document Description
Workshop 0: Foundations Your first serverless deploy — one Lambda + HTTP API with plain CDK, deploy → verify → destroy. Start here if you're new to IaC.
Workshop: Serverless End-to-End Lambda + CDK + AI-DLC + ThothCTL — full application from scratch
Workshop: Enterprise CDK Private constructs, CodeArtifact, Projen, multi-account, RCPs
Workshop: ECS Backend Containers, ECS Express Mode, Docker Compose → cloud, Service Connect
Workshop: CI/CD Phase 1 CDK Pipelines, canary, feature flags, ThothCTL DevSecOps
Workshop: CI/CD Phase 2 Continuum, DevOps Agent, FinOps Agent, supply chain, SCPs + RCPs
Workshop: Micro-Frontends Micro-frontends on CloudFront + S3 vs Amplify, composition, caching

Reference

Document Description
Glossary Definitions of key terms used across the framework, grouped by area, each linking to its dedicated chapter

Key Technologies

Layer Technology Role
AI Methodology AWS AI-DLC + Kiro AI-driven development lifecycle (Inception → Construction → Operations)
Platform CLI ThothCTL DevSecOps automation, scanning, cost analysis, governance
IaC CDK v2 + Express Mode Fast, type-safe infrastructure with compliance (cdk-nag)
Compute Lambda + ECS Fargate Serverless functions + serverless containers
API AppSync + API Gateway GraphQL (real-time) + REST
Events EventBridge Async backbone (Bus + Pipes + Scheduler)
Data DynamoDB + Aurora Serverless v2 NoSQL (default) + SQL (when needed)
AI/ML Bedrock + AgentCore + Strands Foundation models + agent runtime + agent SDK
AI Brain Knowledge Bases + AgentCore Memory + MCP Organizational learning memory for agents + humans
Observability OpenTelemetry (ADOT) + Powertools Vendor-neutral traces/metrics/logs
Security Continuum + Guardrails + cdk-nag Continuous pen testing + AI safety + compliance
Operations DevOps Agent + FinOps Agent Autonomous incident investigation + cost monitoring
Deployment CDK Pipelines + CodeDeploy Self-mutating CI/CD + canary rollouts
Governance SCPs + RCPs + OPA/Rego Principal controls + resource controls + policy-as-code
Scaffold cdkv2_typescript_scaffold Golden path project template

Architecture Principles (Summary)

  1. Serverless-First — Managed services that scale to zero
  2. AI-Native — Bedrock + agents as first-class capabilities
  3. Event-Driven — Async by default; sync only at edges
  4. Cell-Based — Independent, blast-radius-contained units
  5. Hexagonal Code — Ports & adapters; domain never depends on infrastructure
  6. Zero-Trust Security — SCPs + RCPs + Cedar + least privilege
  7. Shift-Left — Security, cost, compliance validated before merge
  8. Observable — OpenTelemetry + Powertools + Application Signals from day one
  9. Cost-Aware — ARM64, SnapStart, Valkey, scales-to-zero
  10. Platform Engineering — Golden paths via constructs, scaffolds, and ThothCTL

Maturity Model

flowchart LR
    L1["L1: TEAM<br/>Single scaffold<br/>Manual deploy<br/>Local cdk-nag"] --> L2["L2: STANDARDIZED<br/>CDK Pipelines<br/>Canary deploy<br/>ThothCTL scan"]
    L2 --> L3["L3: GOVERNED<br/>Private constructs<br/>SCPs + RCPs<br/>OPA policies<br/>Multi-account"]
    L3 --> L4["L4: AUTONOMOUS<br/>AI agents operate<br/>Self-healing<br/>Continuous compliance<br/>FinOps optimization"]
Loading
Level Deploy Security Cost Operations
L1 cdk deploy manually cdk-nag locally Manual review CloudWatch dashboards
L2 CDK Pipelines + canary ThothCTL scan in CI Cost estimation in PR Application Signals
L3 Multi-account + approval gates Continuum + SCPs + RCPs FinOps Agent monitoring DevOps Agent investigation
L4 AI proposes + human approves Auto-remediation Auto-optimization Self-healing with guardrails

Getting Started

Your Goal Start Here
New to AWS IaC — first deploy Workshop 0: Foundations
Build a new serverless app Workshop: Serverless End-to-End
Build a container-based backend Workshop: ECS Backend
Set up enterprise CI/CD Workshop: CI/CD Phase 1
Understand architecture decisions Architecture Principles
Learn software design patterns Software Architecture Patterns
Set up organizational governance Workshop: Enterprise CDK
Build an organizational AI Brain Custom Intelligent AI Brains

View This Documentation Locally

# Install Zensical
pip install zensical

# Preview with hot-reload
zensical serve

# Build static site
zensical build
# → Output in site/ (deploy to S3 + CloudFront, Amplify Hosting, etc.)