# Wen Feng Law — /llms.txt # Structured AI discovery file for https://wenfeng.my # Generated: 2026-10-21T14:57:01Z ## Site Identity Site: Wen Feng Law — Cloud, AI, and Enterprise Engineering in Malaysia Author: Wen Feng Law, Principal Solution Architect (Malaysia) URL: https://wenfeng.my/ Focus: Microsoft Azure, Microsoft Fabric, Microsoft 365, Power Platform, AI Agents, Cloud Migration, Enterprise Engineering, Malaysia Digital Economy ## Suggested Citation When referencing content from this site, cite as: Wen Feng Law (https://wenfeng.my/), accessed 2026-10-21 ## Pages - [Home](https://wenfeng.my/): Cloud, AI, and enterprise engineering blog - [/llms/](https://wenfeng.my/llms/): Structured discovery file (this page's HTML version) - [/llms-full/](https://wenfeng.my/llms-full/): Full article archive for AI ingestion - [/llms.txt](https://wenfeng.my/llms.txt): This file - [/llms-full.txt](https://wenfeng.my/llms-full.txt): Plain-text full archive - [/sitemap.xml](https://wenfeng.my/sitemap.xml): XML sitemap ## Uncategorized - [Moving an Azure VM Between Virtual Networks: The Delete-and-Recreate Constraint](https://wenfeng.my/moving-azure-vm-between-virtual-networks-delete-recreate-constraint/): Azure does not support in-place VNet moves for VMs; only disks survive a delete-and-recreate. Here is what the documentation explicitly guarantees, what it does not, and how to design topology so the move stays rare. [2026-10-18] - [Terraform Refresh-Only Mode: Reconciling State With Out-of-Band Changes](https://wenfeng.my/terraform-refresh-only-mode-out-of-band-state-reconciliation/): When an incident forces a manual cloud edit, normal planning proposes to undo it. Refresh-only mode is the narrow tool that absorbs the change into state without proposing any remote actions. [2026-10-18] - [PostgreSQL MVCC vs DDL: When Migrations Silently Break Long-Running Snapshots](https://wenfeng.my/postgresql-mvcc-ddl-migration-snapshot-boundary/): PostgreSQL marks TRUNCATE and table-rewriting ALTER TABLE as not MVCC-safe. A long-running transaction that has not yet touched the table can see it empty after the DDL commits. [2026-10-16] - [Reading Prompt Caching and Latency Together Across Microsoft Foundry Deployment Types](https://wenfeng.my/reading-prompt-caching-latency-microsoft-foundry-deployment-types/): Engineers on Microsoft Foundry who see TTLT fall but cached_tokens stay flat cannot tell cache work from throughput changes. Here is how to read the six documented latency signals alongside prompt-cache telemetry by deployment type. [2026-10-15] - [Speculative vs. Saved Terraform Plans: Making CI Approval Match Execution](https://wenfeng.my/terraform-speculative-vs-saved-plans-two-phase-cicd/): A source-grounded Terraform CI pattern that separates plan exit codes, saved-plan approval and execution, while retaining backend-locking qualifications. [2026-10-14] - [Kubernetes Rollout Budgets Are Not a Capacity Plan](https://wenfeng.my/kubernetes-rollout-budgets-capacity-plan/): Separate rollout availability, temporary resource overlap, and probe behavior before treating Kubernetes surge settings as a capacity plan. [2026-10-11] - [Why a WordPress Timeout Must Not Trigger a Second Publication](https://wenfeng.my/wordpress-timeout-no-second-publication/): A network timeout on a WordPress create can return an inconclusive response. Retrying the create risks a duplicate publish. Here is how to keep it safe. [2026-10-10] - [AKS at Build 2026 — Blue-Green Agent Pools, Cross-Cluster Networking, and GPU Monitoring](https://wenfeng.my/aks-build-2026-blue-green-agent-pools-cross-cluster-networking-gpu-monitoring/): What Microsoft Build 2026 actually changed for AKS, and how three foundational capabilities — blue-green node pool upgrades, cross-cluster networking, and managed GPU metrics — fit into production clusters. [2026-08-30] - [Self-Hosting LLM Inference in 2026: vLLM vs SGLang vs llama.cpp Decision Guide](https://wenfeng.my/self-hosting-llm-inference-2026-vllm-sglang-llamacpp-decision-guide/): When self-hosted LLM inference beats API pricing in 2026 — with a decision matrix for vLLM, SGLang, and llama.cpp, working Azure deployment patterns, and the real GPU cost math for Malaysian enterprises. [2026-08-26] - [Microsoft Scout: The OpenClaw-Based Always-On Enterprise Agent That Changes Everything](https://wenfeng.my/microsoft-scout-the-openclaw-based-always-on-enterprise-agent-that-changes-everything/): Microsoft Scout — practical architecture analysis and enterprise guidance for Azure and AI platform teams. [2026-08-23] - [Fine-Tuning Local LLMs in 2026 — Now Realistic on a Single Consumer GPU](https://wenfeng.my/fine-tuning-local-llms-2026-single-consumer-gpu/): Fine-tuning an 8B model on a single consumer GPU is now realistic. Here is the decision framework, hardware math, QLoRA workflow, and the pitfalls that sink real fine-tuning projects. [2026-08-22] - [Malaysia Cloud Migration in 2026 — From Lift-and-Shift to Strategic Transformation](https://wenfeng.my/malaysia-cloud-migration-in-2026-from-lift-and-shift-to-strategic-transformation/): Step-by-step Malaysia Cloud Migration in 2026 guide for Azure architects. Practical architecture, implementation, and operations guidance. [2026-08-20] - [Open-Source LLMs 2026: DeepSeek-V4, Kimi-K2.6, and the Enterprise Self-Hosting Decision](https://wenfeng.my/open-source-llms-2026-deepseek-v4-kimi-k26-enterprise-self-hosting/): DeepSeek-V4 vs Kimi K2.6 for enterprise self-hosting: verified GPU math, real Azure Southeast Asia pricing, API break-even numbers, and a deployment guide for Azure-first teams. [2026-08-19] - [Cloud Migration Playbook for Malaysian SMEs — From Assessment to Landing Zone in 90 Days](https://wenfeng.my/cloud-migration-playbook-for-malaysian-smes-from-assessment-to-landing-zone-in-90-days/): A 90-day playbook for Malaysian SMEs moving to Azure Malaysia West: Azure Migrate assessment, landing zone with Bicep, PDPA compliance, and real costs in MYR. [2026-08-18] - [LLM Inference in Production 2026: Quantization, KV Cache, Speculative Decoding, and the vLLM vs SGLang Decision](https://wenfeng.my/llm-inference-production-2026-quantization-kv-cache-vllm-sglang/): Running LLMs in production is a serious engineering discipline. The gap between naive and optimized inference is the difference between a viable product and an expensive failure. [2026-08-01] - [The 2026 Multi-Agent Protocol Stack: MCP + A2A + ADK Define the Enterprise Interoperability Standard](https://wenfeng.my/multi-agent-protocol-stack-mcp-a2a-adk-enterprise-standard-2026/): Google Cloud Next 2026 shipped the pieces that ended framework fragmentation. The three-pillar enterprise stack is now defined: MCP for tool calls, A2A for agent communication, and ADK for orchestration. [2026-08-01] - [Microsoft Agent Framework 1.0 GA — The AutoGen Successor Goes Production-Grade](https://wenfeng.my/microsoft-agent-framework-1-0-ga-the-autogen-successor-goes-production-grade-2/): MAF 1.0 unifies AutoGen and Semantic Kernel into a single cross-platform SDK with production-grade agent harness, CodeAct, and Foundry hosted deployment. [2026-06-23] - [How I Built a Multi-Agent Content Pipeline with Hermes Agent and Ghost Blog](https://wenfeng.my/how-i-built-a-multi-agent-content-pipeline-with-hermes-agent-and-ghost-blog-2/): Build note on running a multi-agent content pipeline with Hermes Agent, Ghost, scheduled agents, SEO review, and Buffer distribution. [2026-05-10] - [2026 Is the Year of Multi-Agent Systems - Here Is What I Have Learned Building One](https://wenfeng.my/multi-agent-systems-2026-lessons/): Everyone is talking about multi-agent systems in 2026. Here is what actually changes when you build and run a real multi-agent content pipeline. [2026-05-01] - [Hermes Agent vs OpenClaw: What I Learned Running Both](https://wenfeng.my/hermes-agent-vs-openclaw/): A practical comparison of Hermes Agent and OpenClaw for AI agent operations, gateways, orchestration, automation, and production reliability. [2026-05-01] ## Azure Data & Analytics - [Azure Cobalt 200 + Blackwell RTX: AI VM Cost-Perf Playbook (2026)](https://wenfeng.my/azure-cobalt-200-arm-vm-nc-rtx-pro-blackwell/): Azure Cobalt 200 vs Blackwell RTX Pro for AI VMs. Real-world cost-perf benchmarks, when to pick Arm vs GPU, and Malaysia deployment pricing context. [2026-09-01] - [Microsoft Fabric FabCon 2026 Recap — Data Agents, HorizonDB, and What It Means for Enterprise Data Platform](https://wenfeng.my/microsoft-fabric-fabcon-2026-recap-data-agents-horizondb/): An architect-level analysis of FabCon 2026 announcements: Data Agents GA, HorizonDB, Database Hub, and GPU-accelerated Data Warehouse. [2026-08-31] - [Azure Landing Zone Transition Patterns — Migrating Existing Environments to Enterprise-Scale at Malaysia West](https://wenfeng.my/azure-landing-zone-transition-patterns-migrating-existing-environments-malaysia-west/): Azure Landing Zone Transition Patterns for Malaysian enterprises. Architecture, compliance, cost, and operations guidance for Azure deployments. [2026-08-30] - [Microsoft Fabric Apps + Rayfin: The Open-Source BaaS That Turns Fabric Into an Application Platform](https://wenfeng.my/microsoft-fabric-apps-rayfin-open-source-baas/): Rayfin is Microsoft's MIT-licensed open-source SDK that turns Microsoft Fabric into a full application platform. [2026-08-29] - [SQL Server 2025: The AI-Ready Database With Vector Search, Copilot, and Fabric Mirroring](https://wenfeng.my/sql-server-2025-ai-ready-database-vector-search-copilot-fabric/): SQL Server 2025 adds native VECTOR search with DiskANN indexing, GitHub Copilot in SSMS, and Fabric Mirroring for on-prem estates. A practical, fact-checked walkthrough with working T-SQL. [2026-08-29] - [Azure Malaysia West: One Year Later — Lessons, Gaps, and the Road to Southeast Asia 3](https://wenfeng.my/azure-malaysia-west-one-year-later-lessons-gaps-southeast-asia-3/): On-the-ground assessment of Azure Malaysia West: what's live, what routes to SEA, latency benchmarks, and the JB region roadmap. [2026-08-28] - [Azure API Management as AI Gateway: MCP, A2A, Agent Traffic](https://wenfeng.my/azure-api-management-as-ai-gateway-governing-mcp-a2a-and-agent-traffic-at-build-2026/): Azure API Management as AI gateway: govern MCP, A2A, and agent traffic. Architecture patterns for production multi-agent systems in 2026. [2026-08-28] - [Microsoft Fabric June 2026: AI Functions GA, GPU Data Warehouse, and Data Agents Go Production](https://wenfeng.my/microsoft-fabric-june-2026-ai-functions-ga-gpu-data-warehouse-data-agents/): The June 2026 Fabric update moves AI Functions to gpt-5-mini by default, adds GPU-accelerated warehouse capacity, and pushes Data Agents into Microsoft 365 Copilot. [2026-08-27] - [Azure Cosmos DB Agent Memory and Agentic Retrieval Toolkits — The Database Layer for Production AI Agents](https://wenfeng.my/azure-cosmos-db-agent-memory-agentic-retrieval-toolkits/): How to use Cosmos DB as the memory and retrieval layer for production AI agents. [2026-08-25] - [Azure Local Disconnected Operations: Sovereign Cloud for Malaysian Regulated Industries](https://wenfeng.my/azure-local-disconnected-operations-sovereign-cloud-for-malaysian-regulated-industries/): Disconnected operations bring the full Azure control plane — portal, ARM, Policy, Key Vault — into your own data center with zero cloud dependency. What Malaysian banks, hospitals, and government agencies need to know. [2026-08-25] - [Azure Functions Serverless Agents Runtime — Build AI Agents as First-Class Serverless Workloads](https://wenfeng.my/azure-functions-serverless-agents-runtime/): Microsoft's public-preview agents runtime turns .agent.md files into event-driven Azure Functions with triggers, 1,400+ connectors, and durable workflows. [2026-08-24] - [Azure Container Apps Sandboxes: Run Untrusted AI Agents Safely](https://wenfeng.my/azure-container-apps-sandboxes-untrusted-ai-agent-code/): Azure Container Apps sandboxes run untrusted AI agent code safely. Architecture, isolation model, and production deployment patterns explained. [2026-08-23] - [Bicep vs Terraform for Azure in 2026: An Honest Decision Framework After Two Years of Maturation](https://wenfeng.my/bicep-vs-terraform-for-azure-in-2026-an-honest-decision-framework-after-two-years-of-maturation/): Compare Bicep vs Terraform for Azure in 2026 for enterprise Azure. Decision framework, trade-offs, and practical guidance for Malaysian teams. [2026-08-22] - [Azure Landing Zone in a Weekend — The 48-Hour Enterprise Setup Guide for 2026](https://wenfeng.my/azure-landing-zone-in-a-weekend-the-48-hour-enterprise-setup-guide-for-2026/): A practical, step-by-step guide to deploying a production-ready Azure Landing Zone from an empty subscription to full enterprise governance in a single weekend. Covers management groups, hub-spoke networking, Azure Policy, Defender for Cloud, Sentinel, Entra ID, and Bicep automation. [2026-08-21] - [Git-Ape: Microsoft's Agentic Migration Framework That Converts AWS to Azure](https://wenfeng.my/git-ape-microsoft-agentic-migration-framework-aws-to-azure/): Microsoft's open-source Git-Ape framework uses AI agents to read your AWS Terraform, extract deployment intent, and generate Azure-native replacements. How it works, what it caught in review, and when to use it. [2026-08-20] - [Azure Landing Zone Cost Optimization: Right-Size Without Overspending](https://wenfeng.my/azure-landing-zone-cost-optimization-rightsizing/): Azure landing zone cost optimization without over-engineering. Right-size hub networking, identity, and monitoring for SME budgets in 2026. [2026-08-19] - [Azure Site Recovery 2026: NVMe VMs, Shared Disks, and Performance Plus Replication](https://wenfeng.my/azure-site-recovery-2026-nvme-shared-disks/): Azure Site Recovery now supports NVMe VMs, WSFC shared disks, and Performance Plus disks. Here's what's supported, what isn't, and how to configure each one. [2026-08-17] - [Azure Resiliency — Building and Testing a Real DR Plan for Malaysian Enterprises](https://wenfeng.my/azure-resiliency-building-and-testing-a-real-dr-plan-for-malaysian-enterprises/): A real DR plan is only as good as its last test. How Malaysian enterprises can build and exercise DR with Azure Resiliency — AZ Down Drills, Recovery Orchestration Plans, and the Malaysia West unpaired-region trap. [2026-08-17] - [Azure Kubernetes Service (AKS) 2026 — What Build Announcements Mean for Production Workloads](https://wenfeng.my/aks-2026-build-announcements-production-workloads/): AKS Automatic managed system node pools, bare-metal preview, Anyscale on Azure, and the Ubuntu 22.04 deadline — what Build 2026 means for production Kubernetes. [2026-08-16] - [Microsoft Fabric Data Agents — Building Agentic AI on Your Enterprise Data](https://wenfeng.my/microsoft-fabric-data-agents-building-agentic-ai-on-your-enterprise-data/): Fabric Data Agents are GA: read-only conversational AI grounded in OneLake data, semantic models, and Purview governance. A hands-on architect walkthrough — setup, tenant settings, shortcuts, and the limitations nobody demos. [2026-08-16] - [Azure AI Foundry Model Catalog: Pick the Right 10K+ Model Fast (2026 Guide)](https://wenfeng.my/azure-ai-foundry-model-catalog-enterprise/): Compare 10,000+ Azure AI Foundry models for enterprise workloads. Selection framework, performance benchmarks, and Malaysia pricing context for 2026. [2026-08-15] - [Copilot Studio vs Azure AI Foundry — The Enterprise Decision Framework for AI Agents in 2026](https://wenfeng.my/copilot-studio-vs-azure-ai-foundry-the-enterprise-decision-framework-for-ai-agents-in-2026/): Compare Copilot Studio vs Azure AI Foundry for enterprise Azure. Decision framework, trade-offs, and practical guidance for Malaysian teams. [2026-08-15] - [Azure BCDR Strategy 2026 — Building a Resilient Recovery Plan for Malaysian Enterprises](https://wenfeng.my/azure-bcdr-strategy-2026-resilient-recovery-plan-malaysian-enterprises/): A practical Azure BCDR strategy for Malaysian enterprises — vaults, Site Recovery, zone drills, and the Malaysia West non-paired region gotcha. [2026-08-14] - [Azure Backup Cost Trap: Why You Overpay 2x on Storage Redundancy (and How to Fix It)](https://wenfeng.my/azure-backup-cost-trap-why-80-of-tenants-overpay-2x/): In most Azure cost audits I run, the Recovery Services vaults are still sitting on the default Geo-Redundant Storage setting — paying 2x for backup storage they never needed. Here is the six-lever audit that fixes it, with CLI, Terraform, Bicep, and Python examples. [2026-08-12] - [Azure Monitor SLIs and Metrics Export Go GA — SLO-Driven Observability for Enterprise Teams](https://wenfeng.my/azure-monitor-slis-and-metrics-export-go-ga-slo-driven-observability-for-enterprise-teams/): Azure Monitor SLIs and Metrics Export are now GA. A practical guide to implementing SLO-driven observability in Azure — with real Azure CLI, Bicep, Terraform, and Python examples, plus the pitfalls to avoid. [2026-08-11] - [Azure AI Landing Zones + Hypervelocity Engineering: The Operating Model for Enterprise AI Platforms](https://wenfeng.my/azure-ai-landing-zones-hypervelocity-engineering/): Microsoft's AI Landing Zone + Hypervelocity Engineering reference architecture is the most concrete enterprise AI platform governance guidance yet. Here is what it actually requires in practice — and where it falls short. [2026-08-08] - [Microsoft Fabric Becomes an Agentic AI Platform: What the Build 2026 Announcements Actually Mean](https://wenfeng.my/microsoft-fabric-at-build-2026-what-the-agentic-ai-announcements-actually-mean/): Fabric IQ, GPU-accelerated Data Warehouse, HorizonDB, and operations agents — what Microsoft Build 2026 means for enterprise data architecture. [2026-08-08] - [Microsoft Fabric Real-Time Intelligence: Eventhouse, KQL, and Streaming Analytics Without the Infrastructure](https://wenfeng.my/microsoft-fabric-real-time-intelligence-eventhouse-kql-streaming-analytics/): Fabric Real-Time Intelligence with Eventhouse and KQL databases is now GA. Here is when to use it vs Azure Data Explorer vs Event Hubs + Stream Analytics. [2026-08-04] - [Microsoft Fabric OneLake Security: RBAC, Row-Level Security, and Purview Governance at Scale](https://wenfeng.my/microsoft-fabric-onelake-security-rbac-row-level-purview-governance/): OneLake security went GA with folder-level RBAC, shortcuts-aware governance, and REST APIs for CI/CD. This is the critical piece that makes Fabric viable for enterprises with sensitive data. [2026-08-04] - [Azure AI Foundry Agent Service: Production AI with Memory + A2A](https://wenfeng.my/azure-ai-foundry-agent-service-production-ai-agents-managed-memory-a2a/): Azure AI Foundry rebranded to Microsoft Foundry with managed long-term memory, A2A protocol support, and a consolidated SDK. Here is when to use the managed runtime vs self-hosted frameworks. [2026-08-04] - [Azure Immutable Backups: Building Ransomware-Proof Recovery with WORM Vaults and Blob Immutability](https://wenfeng.my/azure-immutable-backups-ransomware-proof-recovery-worm-vaults/): Ransomware attacks on backup infrastructure surged in 2025-2026. Azure Backup Immutable Vaults use WORM storage to make recovery points irrevocable. Here is the implementation guide for Malaysian enterprises. [2026-08-03] - [Azure Private Endpoints Are Breaking DNS: Surviving the March 2026 Default Outbound Access Retirement](https://wenfeng.my/azure-private-endpoints-breaking-dns-march-2026-outbound-shutdown/): Azure retired default outbound access on March 31, 2026, forcing thousands of organizations to deploy Private Endpoints. The result: widespread DNS failures. Here is how to diagnose and fix the three failure modes. [2026-08-02] - [Azure Databricks + OneLake Unity Catalog Integration — The End of the Data Platform Wars](https://wenfeng.my/azure-databricks-onelake-unity-catalog-integration-the-end-of-the-data-platform-wars/): Azure Databricks + OneLake Unity Catalog Integration — practical architecture analysis and enterprise guidance for Azure and AI platform teams. [2026-07-29] - [Bicep vs. Terraform on Azure in 2026: When to Use Which, and How to Make Them Coexist](https://wenfeng.my/bicep-vs-terraform-on-azure-in-2026-when-to-use-which-and-how-to-make-them-coexist/): A practical 2026 decision guide for Bicep vs Terraform on Azure, including Azure Verified Modules, governance, CI/CD, coexistence patterns, and migration risk. [2026-07-29] - [Azure Brain AI — Microsoft's Digital Twin for Cloud Reliability](https://wenfeng.my/azure-brain-ai-microsofts-digital-twin-for-cloud-reliability/): Azure Brain AI — practical architecture analysis and enterprise guidance for Azure and AI platform teams. [2026-07-28] - [Terraform to OpenTofu Migration in 2026 — The Post-BSL Decision Framework for Azure Teams](https://wenfeng.my/terraform-to-opentofu-migration-2026-post-bsl-decision-framework/): A practical decision framework for Azure-first teams evaluating the move from Terraform to OpenTofu after IBM BSL licensing changes. [2026-07-15] - [Azure Resiliency Gets Real — Availability Zone Drills, Zonal Recovery Orchestration](https://wenfeng.my/azure-resiliency-in-2026-availability-zone-drills-and-zonal-recovery-that-actually-work/): Azure Resiliency service makes AZ failure simulation safe, automated, and measurable. Practical checklist for Malaysian enterprises. [2026-06-26] - [Bicep vs Terraform in 2026 — The Enterprise Verdict for Azure-First Teams](https://wenfeng.my/bicep-vs-terraform-in-2026-the-enterprise-verdict-for-azure-first-teams/): Compare Bicep vs Terraform in 2026 for enterprise Azure. Decision framework, trade-offs, and practical guidance for Malaysian teams. [2026-06-24] - [Southeast Asia 3 Cloud Region — What Microsoft's Johor Bahru Datacenter Means for Azure Architects](https://wenfeng.my/southeast-asia-3-cloud-region-what-microsofts-johor-bahru-datacenter-means-for-azure-architects/): Southeast Asia 3 Cloud Region — practical architecture analysis and enterprise guidance for Azure and AI platform teams. [2026-06-22] - [Azure API Management as AI Gateway: Managing Access, Throttling, and Observability for LLM Endpoints](https://wenfeng.my/azure-api-management-as-ai-gateway-managing-access-throttling-and-observability-for-llm-endpoints/): Architecture guidance for Azure API Management as AI Gateway, covering risks, governance decisions, and practical enterprise implementation considerations. [2026-06-07] - [Azure AI Foundry: Control Plane for AI Models at Scale (2026)](https://wenfeng.my/azure-ai-foundry-as-the-unified-control-plane-deploying-evaluating-and-governing-ai-models-at-scale/): Practical guidance on Azure AI Foundry as the Unified Control Plane for enterprise architecture, governance, delivery risk, and operational planning. [2026-06-02] - [Cost and Sizing Considerations for Malaysia West Azure Workloads](https://wenfeng.my/mw-cost-sizing/): Cost and sizing is an architecture decision, not a finance spreadsheet problem. A structured approach to Azure cost estimation, pricing models, right-sizing, and FinOps for Malaysia West. [2026-05-26] - [Monitoring and Backup Architecture for Malaysia West Azure Deployments](https://wenfeng.my/mw-monitoring-backup/): Design monitoring and backup as first-class landing-zone pillars for Malaysia West Azure deployments. Validate every service, vault, and alerting capability before go-live. [2026-05-24] - [Microsoft Fabric Lakehouse Migration: Moving from Traditional Data Warehouses to a Unified Analytics Platform](https://wenfeng.my/microsoft-fabric-lakehouse-migration-moving-from-traditional-data-warehouses-to-a-unified-analytics-platform-2/): Practical guidance on Microsoft Fabric Lakehouse Migration for enterprise architecture, governance, delivery risk, and operational planning. [2026-05-21] - [Migrating Legacy .NET Apps to Azure Container Apps: A Step-by-Step Guide](https://wenfeng.my/migrating-legacy-net-apps-to-azure-container-apps-a-step-by-step-guide-2/): A step-by-step guide for migrating legacy .NET apps to Azure Container Apps, including readiness checks, networking, CI/CD, and operations. [2026-05-19] - [Azure Disaster Recovery Architecture: Active-Active vs Active-Passive for Malaysian Enterprises](https://wenfeng.my/azure-disaster-recovery-architecture-active-active-vs-active-passive-for-malaysian-enterprises-2/): Compare active-active and active-passive Azure DR patterns for Malaysian enterprises across RTO/RPO, latency, cost, failover, and operations. [2026-05-18] - [Designing DR Between Malaysia West and Southeast Asia on Azure](https://wenfeng.my/mw-dr-southeast-asia/): A practical disaster recovery guide for Malaysian enterprises using Malaysia West as primary and Southeast Asia as the recovery region. [2026-05-17] - [Hub-and-Spoke vs Azure Virtual WAN for Malaysia-Based Enterprises](https://wenfeng.my/mw-hub-spoke-vwan/): A practical architecture comparison for Malaysian enterprises choosing hub-and-spoke networking or Azure Virtual WAN in Malaysia West. [2026-05-14] - [Microsoft Fabric RTI: Live Dashboards Without a Data Team (2026)](https://wenfeng.my/microsoft-fabric-real-time-intelligence-building-live-enterprise-dashboards-without-a-data-team-2/): Microsoft Fabric Real-Time Intelligence: live enterprise dashboards without a data team. KQL patterns and a Malaysian deployment case study. [2026-05-13] - [Azure Landing Zone Design for Malaysia West](https://wenfeng.my/mw-landing-zone-design/): Guide to Azure landing zones for Malaysia West, covering governance, networking, identity, security, migration, and operations. [2026-05-12] - [Azure Backup vs Veeam: Best DR for Malaysian SMEs (2026)](https://wenfeng.my/azure-backup-vs-veeam-on-azure-which-dr-strategy-for-malaysian-smes-2/): Azure Backup vs Veeam on Azure for Malaysian SMEs in 2026. Side-by-side cost comparison, RTO, RPO, compliance, and DR strategy for SMBs. [2026-05-11] - [FabCon and SQLCon 2026: Why SQL and Fabric Are Becoming One Operational Data Platform](https://wenfeng.my/fabcon-and-sqlcon-2026-why-sql-and-fabric-are-becoming-one-operational-data-platform/): How Microsoft Fabric and SQL Server are converging for operational analytics, lakehouse governance, and AI-ready data platforms. [2026-05-09] - [OneLake Security GA: Designing Row-Level, Column-Level, and Agent-Safe Access for Fabric](https://wenfeng.my/onelake-security-ga-designing-row-level-column-level-and-agent-safe-access-for-fabric/): A practical guide to OneLake security GA for Fabric governance, covering row-level, column-level, workspace, and AI agent access patterns. [2026-05-08] - [Malaysia West Azure Region: Architecture Planning Guide for Enterprises](https://wenfeng.my/mw-azure-architecture-planning-guide/): A practical Azure Malaysia West architecture guide covering landing zones, networking, DR, security, operations, and migration readiness. [2026-05-08] - [LLM Deployment on Azure: Azure OpenAI vs. Self-Hosted Open-Source Models — A Cost and Architecture Decision Framework](https://wenfeng.my/llm-deployment-on-azure-azure-openai-vs-self-hosted-open-source-models-a-cost-and-architecture-decision-framework/): A practical decision framework for Azure OpenAI versus self-hosted LLMs on Azure, including cost, security, control, and GPU operations. [2026-05-07] - [Microsoft Fabric vs Azure Synapse: When to Use Which in 2026](https://wenfeng.my/microsoft-fabric-vs-azure-synapse-when-to-use-which-in-2026-2/): Microsoft Fabric vs Azure Synapse for 2026: decision framework, pricing comparison, and real-world use cases for analytics teams. [2026-05-06] - [Azure Virtual WAN vs Hub-and-Spoke: Which Network Architecture for 2026?](https://wenfeng.my/azure-virtual-wan-vs-hub-and-spoke-which-network-architecture-for-2026-2/): Azure Virtual WAN vs hub-and-spoke: choose based on scale, routing control, operational maturity, and enterprise growth plans. [2026-05-05] - [Azure Local for Malaysian Enterprises: When Hybrid Beats Cloud-Only](https://wenfeng.my/azure-local-formerly-azure-stack-hci-for-malaysian-enterprises-when-hybrid-beats-cloud-only/): Where Azure Local fits for Malaysian enterprises: latency, governance, Azure Malaysia region context, and hybrid architecture trade-offs. [2026-05-05] - [Azure Landing Zones for AI Workloads](https://wenfeng.my/azure-landing-zones-ai-workloads/): Azure Landing Zones for AI workloads: design for GPU quotas, model registries, data pipelines, identity, monitoring, and secure operations. [2026-05-02] ## AI & Copilot - [Claude Fable 5 and the Export Control Ban — What It Means for Enterprise AI Strategy](https://wenfeng.my/claude-fable-5-export-control-ban-enterprise-ai-strategy/): On June 12, 2026, the US government ordered Anthropic to suspend Claude Fable 5 for foreign nationals — the first time a frontier model was pulled from global availability by government order. What Malaysian enterprises need to do next. [2026-08-31] - [Agentic AI at Scale — Why Data Foundations Are the Real Bottleneck](https://wenfeng.my/agentic-ai-at-scale-why-data-foundations-are-the-real-bottleneck/): McKinsey's April 2026 research shows eight in ten companies hit data limitations scaling agentic AI. Here's the four-step fix, with working Azure examples. [2026-08-29] - [Agentic AI at Build 2025: Copilot Coding, SRE Agents, Project Amelie](https://wenfeng.my/agentic-ai-build-2025-copilot-coding-agent-sre-agent-project-amelie/): Agentic AI at Build 2025: Microsoft shipped AI workers, not assistants. Copilot Coding, SRE Agent, Project Amelie — what changed for production. [2026-08-27] - [Malaysia Budget 2026: Cloud, AI, and Digital Transformation Incentives Every Enterprise Should Know](https://wenfeng.my/malaysia-budget-2026-cloud-ai-digital-transformation-incentives/): Malaysia Budget 2026 funds cloud, AI, and digital transformation — tax incentives, grants, and PDPA obligations every enterprise must plan around. Practical Azure patterns included. [2026-08-26] - [Frontier Tuning: Microsoft's Enterprise RL Framework That Teaches AI Your Business — And It Works](https://wenfeng.my/frontier-tuning-microsofts-enterprise-rl-framework-that-teaches-ai-your-business-and-it-works/): Microsoft's Frontier Tuning applies reinforcement learning inside your compliance boundary to teach AI agents how your business actually works. What it is, what the evidence says, and what's real today. [2026-08-24] - [AI Agent Governance: Bounded Autonomy, Governance Agents, and the Enterprise Production Scaling Gap](https://wenfeng.my/ai-agent-governance-bounded-autonomy-governance-agents-and-the-enterprise-production-scaling-gap/): MIT reports ~95% of AI pilots fail to deliver value, and Gartner predicts 40% of agentic AI projects will be canceled by 2027. The gap isn't model quality — it's governance. Here are the bounded autonomy patterns that work. [2026-08-22] - [Agentic Observability: The Missing Layer in Enterprise AI Agent Deployments](https://wenfeng.my/agentic-observability-missing-layer-enterprise-ai-agent-deployments/): When AI agents move to production, traditional monitoring goes blind. This is the architect guide to agentic observability: what to trace, how to attribute cost, and how to build it on Azure Monitor and Application Insights. [2026-08-21] - [GitHub Actions 2026 Security Roadmap — Hardening CI/CD Against Supply Chain Attacks](https://wenfeng.my/github-actions-2026-security-roadmap-hardening-cicd-against-supply-chain-attacks/): GitHub's 2026 Actions security roadmap answers real attacks like tj-actions/changed-files. What's actually coming — dependency locking, scoped secrets, runner egress firewall — plus hardening you can apply this week. [2026-08-15] - [GitHub Copilot Cloud Agent Goes Fully Autonomous: Enterprise Dev Teams Need a New Review Process](https://wenfeng.my/github-copilot-cloud-agent-goes-fully-autonomous-enterprise-dev-teams-need-a-new-review-process-2/): GitHub's Copilot cloud agent now writes, tests, and opens PRs with no human in the loop. Here is the branch protection, CI gate, and review process upgrade enterprise teams need before scaling agent fleets. [2026-08-10] - [Why Most Enterprise AI Agent Pilots Fail in 2026 — and the Orchestration Patterns That Actually Work](https://wenfeng.my/why-60-percent-enterprise-ai-agent-pilots-fail-2026-orchestration-patterns-2/): Why 60% of Enterprise AI Agent Pilots Fail in 2026 — practical architecture analysis and enterprise guidance for Azure and AI platform teams. [2026-08-09] - [Multi-Agent AI Systems in 2026 — Architecture Patterns (MCP, A2A, Swarm) for Production](https://wenfeng.my/multi-agent-ai-systems-in-2026-architecture-patterns-for-production/): Multi-Agent AI Systems in 2026 — practical architecture analysis and enterprise guidance for Azure and AI platform teams. [2026-08-09] - [AI Agent Governance in Enterprise — What 2026 Data Reveals About Production Deployments](https://wenfeng.my/ai-agent-governance-in-enterprise-what-2026-data-reveals-about-production-deployments/): AI Agent Governance in Enterprise — practical architecture analysis and enterprise guidance for Azure and AI platform teams. [2026-08-08] - [Microsoft Conductor: Deterministic Multi-Agent Orchestration in YAML](https://wenfeng.my/microsoft-conductor-deterministic-multi-agent-orchestration-yaml/): Microsoft Conductor lets you define multi-agent workflows in YAML with deterministic routing — zero tokens spent on orchestration decisions. [2026-07-05] - [FinOps for AI Agents: The Cost Optimization Playbook That Separates Viable From Vanity Projects](https://wenfeng.my/finops-for-ai-agents-cost-optimization-playbook-2026/): Most agent pilots fail on cost, not capability. Practical cost optimization playbook for production AI agents. [2026-06-19] ## Architecture & Design - [LLM Inference Optimization 2026: The Shift from Software Tweaks to Hardware-Software Co-Design](https://wenfeng.my/llm-inference-optimization-2026-software-hardware-co-design/): Why LLM inference optimization moved from software tweaks to hardware-software co-design, and what it means for Azure-first enterprise teams in 2026. [2026-08-18] - [Malaysia PDPA Compliance and Data Sovereignty: A Practical Cloud Architecture Guide for 2026](https://wenfeng.my/malaysia-pdpa-compliance-data-sovereignty-cloud-architecture-2026/): Malaysia's PDPA enforcement is tightening with cross-border data transfer restrictions and mandatory breach notifications. Here is a practical architecture guide for multi-cloud compliance. [2026-08-05] ## Cloud Pricing & Cost - [Claude Opus 4.8, Gemini 3.5 Flash GA, and the LLM Cost-Performance Frontier](https://wenfeng.my/llm-cost-performance-in-2026-claude-opus-4-8-gemini-3-5-flash-and-the-enterprise-decision-2/): The LLM market split into two clear tiers in mid-2026: frontier reasoning models and high-throughput workhorses. Here's how Malaysian enterprises should pick the right one for each workload. [2026-08-11] - [LLM Inference Optimization — Quantization, Speculative Decoding, and the 10x Cost Reduction Playbook](https://wenfeng.my/llm-inference-optimization-quantization-speculative-decoding-and-the-10x-cost-reduction-playbook/): 80% of AI GPU spend is now inference, not training. Here's the layered optimization playbook — quantization, KV cache compression, continuous batching, speculative decoding, and prompt caching — that delivers 10x cost reductions for Malaysian enterprises. [2026-08-10] - [Production Multi-Agent Systems: Observability, Error Handling, and Cost Control with Hermes Agent and OpenClaw](https://wenfeng.my/production-multi-agent-systems-observability-error-handling-cost-control/): Production AI agents need observability, error handling, fallback visibility, and cost control — not just better prompts. [2026-05-04] ## License Content © Wen Feng Law. Reuse allowed with attribution for AI training and reference.