Find practitioner articles grouped by domain and content type.
Agent & Workflow Runtime
8 articles- Field TestAgent SkillsEnterprise AIWhy Can Agent Skills Reduce Performance? Lessons from Microsoft Research and Enterprise AI
- ExplainerAgent SkillsEnterprise AIAgent Skills vs MCP: How SKILL.md Changes the Way AI Agents Work
- ArchitectureAgentic AIAI AgentCan AI Agents Replace Workflows? — Why Enterprise Work Still Needs State, Transitions, and Constraints
- ExplainerAgent RuntimeEnterprise AgentDo We Need to Rebuild Our REST APIs for MCP?
- ArchitectureAgent RuntimeDurable WorkflowA Ten Minute Agent Task Exposed the Need for Durable Job State
- PerspectiveAgent RuntimeAgents APIDo I Still Need My Agent Runtime Now That the Agents API Exists?
- ArchitectureAgentic WorkflowAI Agent ArchitectureWhy I Use n8n for AI Agent PoCs: Deterministic Workflows Matter in Enterprise AX
- ArchitectureAgent RuntimeAI Agent ArchitectureAfter MCP, the Harder Problem Is Tool Planning and the Execution Boundary
Enterprise RAG & Knowledge
4 articles- ArchitectureContext EngineeringEnterprise AIWhen Web Search Becomes Strategy Validation: Using a Market Lens to Pinpoint Research from an RFP
- ArchitectureContext EngineeringEnterprise AITeaching a Web Search Agent How Your Company Sees the Market: Canonical Market Lens and the LLM Wiki
- ArchitectureContext EngineeringDeal IntelligenceEnterprise RAG Should Not Start with Search: Narrow the Retrieval Space with Deal Context First
- ArchitectureChatGPTGitHubWhy I Don’t Leave Project State to ChatGPT Memory: Building a Personal LLM Wiki with GitHub and Obsidian
CRM & Decision Automation
7 articles- Field TestJEVMicrosoft Decision-1Why Did the Same Decision API Produce Different Answers? Question Engineering with Microsoft Decision-1 and JEV
- Field TestAI ClassificationGPT-6 LunaJEV and GPT-6 Luna Agreed on All 17. So Why Did the Designs Differ?
- Field TestAI ConfidenceCalibrationAI Says It’s 90% Confident. Should You Automate?
- Field TestDecision HarnessEmail ClassificationWhat Building an Email Classifier with JEV Taught Me About Decision Boundaries
- ArchitectureAI ClassificationAI EvaluationSalesforce Email-to-Case AI Classification ③ Accuracy Is Built in Operations
- ArchitectureAI ClassificationEmail-to-CaseSalesforce Email-to-Case AI Classification ② What Happens When Intent Changes Inside the Same Email Thread?
- ArchitectureAI ClassificationEmail-to-CaseSalesforce Email-to-Case AI Classification ① What I Had to Understand Before Automating It
Voice AI & Contact Center
3 articles- ArchitectureAgent RuntimeContact CenterFrom B2B Voice Agents to an AI Call Center: Designing Campaign Journeys That Remember the Customer
- ArchitectureAgent RuntimeAI GovernanceWhat Should a Voice Agent Verify Before Calling a Customer? — Security, Consent, and Do-Not-Call Boundaries for Enterprise Voice AI
- ArchitectureContact CenterGPT-LiveHow Should an Enterprise Design a Voice Agent? — From Callbots to GPT-Live, What Changed After Building Them
AI Engineering & DevOps
4 articles- PerspectiveAI Code ReviewCoding AgentsThe AI Code Review Bottleneck: Why Does Faster Code Generation Create More Work?
- Field TestAI CodingAntigravityCodex Built It. Antigravity Reviewed It — Coding Agents Need Role Separation Too
- ArchitectureAgent Execution HarnessAI Agent ArchitectureWhy I Put Codex CLI Behind an API: Agent Runtime Design for Faster AI PoCs
- ArchitectureAI AutomationChatGPTWhen ChatGPT Commits, the Server Moves: AI Automation with GitHub, n8n, and OCI
AX Strategy & Adoption
4 articles- ArchitectureAI PoCAI ROIHow Should Enterprises Calculate AI ROI? Measure Cost per Accepted Business Outcome, Not Token Spend
- PerspectiveAI AgentAI GovernanceWhat Enterprise AX Consultants Actually Do: Design Boundaries Before Choosing AI
- PerspectiveAI AdoptionAXThe Hardest Part of AI Adoption Was Trust: Lessons from Silent Resistance
- PerspectiveAgentic AIContext EngineeringThe 95% AI Pilot Failure Number Is Not the Real Story