Cloud & Infrastructure

Where the code actually runs.

AWS, Kubernetes, containers, CI/CD, observability and platform reliability.

31 posts · All topics

5

Part 5 — Mem0 Intro Part 5 — 生產部署 — OSS 與 Platform 的分界線在哪裡

Mem0 原始碼導讀系列最終篇：拆解 OSS 與 Platform 的完整差異、Graph Memory／Temporal Reasoning／Memory Decay／Dream 四個專屬能力各自解決什麼問題與 OSS 的補法、v2→v3 破壞性變更遷移清單、怎麼用官方開源評測框架測自己的資料，以及一份生產檢查清單。

·27 min aiengineeringinfrastructurearchitecture
4

Part 4 — Mem0 Intro Part 4 — 儲存層與後端選型 — 三個 store、25 種向量庫與那個關鍵能力差異

Mem0 原始碼導讀系列第四篇：拆解向量庫／實體庫／SQL 三個儲存層的實際 schema、25 種向量庫裡哪 15 種真的支援 BM25、配置系統的組裝方式、LLM 與 embedder 的選型判準，以及自架 server 的部署與多租戶隔離。

·26 min aiengineeringinfrastructure
5

Part 5 — vLLM Intro Part 5 — 生產部署與服務化 — 從 vllm serve 到一份可被驗收的 SLO

vLLM 原始碼導讀系列最終篇：拆解 OpenAI 相容 API 的完整面、Multi-LoRA 多租戶服務、結構化輸出的取樣層約束、Prometheus 指標全表與症狀診斷鏈、P/D 分離與 KV Connector 的實際配置，以及一套不會騙自己的 benchmark 方法。

·28 min aiengineeringinfrastructurearchitecture
4

Part 4 — vLLM Intro Part 4 — 分散式推論、量化與編譯優化 — 讓模型放得下也跑得快

vLLM 原始碼導讀系列第四篇：拆解 TP／PP／DP／EP 四種平行度的切法與通訊量、多機部署與 NCCL 排錯、量化方法與硬體支援矩陣、torch.compile 的 piecewise CUDA Graph 為什麼切在 attention 上，以及 attention backend 的選擇邏輯。

·27 min aiengineeringinfrastructure
1

Part 1 — vLLM Intro Part 1 — 全景架構 — 從一次 model.generate() 到一個推論引擎

vLLM 原始碼與架構導讀系列第一篇：用一張全景圖說清楚這個 9 萬星的 LLM 推論引擎由哪些子系統組成、一個請求進來之後經過哪些階段、GPU 記憶體到底被誰吃掉，以及 V1 重寫改變了什麼。

·24 min aiengineeringinfrastructure
3

Part 3 — 用 AI Bot 打造顧問團隊（三）：評估、維運與優化計畫

AI 顧問 Agent 團隊上線後怎麼辦？本文從 DevOps/SRE 角度，涵蓋系統效能評估、品質驗證、監控告警、部署策略、以及持續改善的 Roadmap。

·25 min aiinfrastructure

Harness 工程入門指南：AI 時代的基礎設施自動化

深入探討 Harness 在 AI 時代的角色，從基本概念、核心功能到實戰應用，幫助工程團隊建立高效的自動化部署流程，加速 AI 應用的上線速度。

·35 min aiinfrastructure

Fine-Tuning LLMs with AWS Bedrock: A Complete Guide to Post-Training Customization

Comprehensive guide to fine-tuning and customizing Large Language Models (LLMs) with AWS Bedrock - covering supervised fine-tuning, continued pre-training, and reinforcement fine-tuning with practical examples and AWS CDK infrastructure setup.

·28 min aiinfrastructure

Building a Centralized Monitoring System with AWS CloudWatch and Grafana using CDK

Design and implement a production-ready centralized monitoring and observability platform using AWS CloudWatch, Grafana, and CDK (TypeScript). Learn cross-service log aggregation, metric collection, IAM permissions, dashboard creation, and alerting strategies for monitoring distributed systems.

·65 min aiengineeringinfrastructure

Building a Centralized User Access Control System with AWS Cognito and CDK

Design and implement a production-ready centralized authentication and authorization system using AWS Cognito and CDK (TypeScript). Learn multi-tenant strategies, user pool management, RBAC/ABAC patterns, and how to integrate with multiple services and applications.

·60 min aiengineeringarchitectureinfrastructure

Deploying Hugging Face Models to AWS: A Complete Guide with CDK, SageMaker, and Lambda

Learn how to deploy production-ready Hugging Face AI models to AWS using CDK (TypeScript), SageMaker, and Lambda. Comprehensive guide covering system design, infrastructure setup, model deployment, API creation, and best practices for scalable ML applications.

·55 min aiengineeringinfrastructure
8

Part 8 — Kubernetes Autoscaling Complete Guide (Part 8): Security, Compliance & Governance

Part 8 of the Kubernetes Autoscaling series: Complete guide to securing autoscaling infrastructure with RBAC, policy enforcement, compliance frameworks (PCI-DSS, HIPAA, SOC2), multi-tenancy patterns, audit logging, and governance best practices for enterprise Kubernetes.

·50 min engineeringinfrastructure
7

Part 7 — Kubernetes Autoscaling Complete Guide (Part 7): Production Troubleshooting & War Stories

Part 7 of the Kubernetes Autoscaling series: Real-world production incidents, debugging workflows, common failure scenarios, and hard-learned lessons from operating autoscaling at scale. Battle-tested troubleshooting guides and postmortem analysis.

·45 min engineeringinfrastructure
6

Part 6 — Kubernetes Autoscaling Complete Guide (Part 6): Advanced Autoscaling Patterns

Part 6 of the Kubernetes Autoscaling series: Advanced autoscaling patterns for stateful applications, multi-cluster deployments, cost optimization strategies, batch job scaling, and emerging technologies. Real-world architectures and production-grade implementations.

·40 min engineeringarchitectureinfrastructure
5

Part 5 — Kubernetes Autoscaling Complete Guide (Part 5): Vertical Pod Autoscaler & Resource Optimization

Part 5 of the Kubernetes Autoscaling series: Deep dive into Vertical Pod Autoscaler (VPA), resource right-sizing strategies, combining VPA with HPA, and production-grade resource optimization techniques for cost-effective Kubernetes operations.

·35 min engineeringinfrastructure
4

Part 4 — Kubernetes Autoscaling Complete Guide (Part 4): Monitoring, Alerting & Threshold Tuning

Part 4 of the Kubernetes Autoscaling series: Complete guide to monitoring EKS autoscaling with Prometheus and Grafana. Includes CDK setup, alerting rules, custom dashboards, and threshold tuning strategies for production-grade observability.

·30 min engineeringinfrastructure
3

Part 3 — Kubernetes Autoscaling Complete Guide (Part 3): Hands-On HPA Demo with Apache-PHP

Part 3 of the Kubernetes Autoscaling series: Hands-on tutorial demonstrating Horizontal Pod Autoscaler with a real Apache-PHP application. Includes complete AWS CDK infrastructure code, Kubernetes manifests, load testing, and step-by-step deployment guide.

·25 min engineeringinfrastructure
2

Part 2 — Kubernetes Autoscaling Complete Guide (Part 2): Cluster Autoscaling & Cloud Providers

Part 2 of the Kubernetes Autoscaling series: Comprehensive guide to cluster-level autoscaling covering Cluster Autoscaler, Karpenter, cloud provider-specific solutions (EKS, GKE, AKS), and emerging technologies for intelligent node provisioning and cost optimization.

·32 min engineeringinfrastructure
1

Part 1 — Kubernetes Autoscaling Complete Guide (Part 1): Horizontal Pod Autoscaler

Part 1 of the Kubernetes Autoscaling series: Deep dive into Horizontal Pod Autoscaler (HPA) approaches, comparing resource-based, custom metrics, external metrics, and event-driven autoscaling with KEDA. Learn when to use each approach with real-world examples and production best practices.

·28 min engineeringinfrastructure

Spring Boot 多環境配置完整指南：開發、測試、生產環境管理

深入探討 Spring Boot 多環境配置管理，包括資料庫切換、Redis 配置、以及 Docker 容器化部署的完整實作指南。

·18 min engineeringinfrastructure
3

Part 3 — Kubernetes 完整指南（三）：進階功能與生產環境實踐

深入探討 Kubernetes 進階主題，包含自動擴展、RBAC 權限管理、Network Policy、Helm 套件管理、監控告警、日誌收集、CI/CD 整合與生產環境最佳實踐，打造企業級 K8S 平台。

·75 min engineeringinfrastructure
2

Part 2 — Kubernetes 完整指南（二）：核心資源與 kubectl 實戰操作

深入探討 Kubernetes 核心資源對象，包含 Pod、Deployment、Service、Ingress、Volume 等完整操作指南，搭配大量 kubectl 指令範例與 YAML 配置，從基礎到實戰全面掌握。

·70 min engineeringinfrastructure
1

Part 1 — Kubernetes 完整指南（一）：基礎概念與架構詳解

深入淺出介紹 Kubernetes 容器編排平台，涵蓋核心概念、架構設計、元件功能、與 Docker 的關係，以及完整的安裝配置教學。從零開始掌握 K8S 基礎知識。

·60 min engineeringinfrastructure
3

Part 3 — Docker 完整指南（三）：進階應用與生產實踐

深入探討 Docker 進階主題，包含 Dockerfile 最佳實踐、多階段建立、Docker Compose 編排、網路進階配置、安全性強化、效能優化與生產環境部署策略。

·70 min engineeringinfrastructure
2

Part 2 — Docker 完整指南（二）：指令大全與實務操作

完整的 Docker 指令參考手冊，涵蓋容器管理、映像操作、網路配置、資料卷管理等實務操作，配合大量範例與表格說明，從基礎到進階全面掌握 Docker CLI。

·60 min engineeringinfrastructure