Writing
Articles and field notes on AI systems, governance, decisions, architecture, compliance automation, and production AI. Practical writing from enterprise implementation experience.
Where Should Your Company Use AI? A Decision Framework
A practical framework for identifying AI opportunities worth pursuing, separating business problems from technology ideas, and deciding what should remain human before committing to production.
AI use case discoveryClaude Code MCP Failed to Connect: stdio, ENOENT, Timeouts
Claude Code MCP server failed to connect? The error is a status, not a root cause. Diagnose pending approval, ENOENT, stdio startup, stdout corruption, timeouts, the 2026-07-28 protocol change, remote gateway failures, and secure database access boundaries in order.
Low-Latency AI Voice Pipeline: Modal, FastAPI & Postgres
modal fastapi voice agentFix FastAPI WebSocket 1006 During LLM Streaming: Timeouts, Async Blocking, Keepalive
FastAPI WebSocket closing with code 1006 during slow LLM generation? The problem is rarely a single timeout. Diagnose event-loop blocking, Uvicorn ping settings, proxy idle limits, and keepalive architecture correctly.
Fix vLLM CUDA Out of Memory at Startup: KV Cache + Docker
vLLM CUDA out of memoryTexas HB 149 vs EU AI Act: A CTO's Engineering Compliance Playbook
Texas HB 149 and the EU AI Act are structurally different laws that create surprisingly similar engineering problems: AI inventory, CI/CD gates, evidence generation, and drift monitoring.
12 Production Readiness Checks for AI Pilots: An Enterprise Architecture Gate
12 production readiness checks for enterprise AI pilots. Each check maps to a specific failure mode, architectural control, and evidence artifact required for audit.
AI production readinessRecover a Stalled AI Pilot in 30 Days
Implementing Immutable Audit Trails for SOC 2 AI Compliance
This guide outlines practical steps to implement immutable audit trails for SOC 2 compliance, enhancing data integrity and security.
immutable audit trailsThe Hidden SEO Risk in AI-Assisted Frontend Development
A technical whitepaper examining how client-only rendering in AI-assisted Next.js and React apps reduces crawlability, delays Google indexing, and obscures public web content with audit checklist and server-first patterns.
Build a Logging Pipeline for EU AI Act Compliance
EU AI ActWhat Is llms.txt? Why Every AI-Ready Website Should Publish It
A builder's case for llms.txt, llms-full.txt, and the emerging agentic web with evidence, implementation templates, and a real retrieval case study.
HAIEC: A Modular AI Governance Framework Explained
A modular AI governance platform built for EU AI Act, NIST AI RMF, and ISO 42001 compliance. Four core modules: Compliance Engine, Red Audit Kit, Precision Drift Detection, and LegacyShift.
AI governanceHow AI Voice Agents Work: KestrelVoice Architecture detailed look
How KestrelVoice works under the hood: telephony, adaptive orchestration, RAG, tool execution, degradation modes, security, and the evidence pipeline that makes AI voice agents production-ready.
Implementing RAG Row-Level Security for Multi-Tenant AI
RAG row-level securityProduction RAG Architecture Patterns for Hybrid Search
Implement hybrid search for production RAG: BM25 lexical plus vector semantic search with reciprocal rank fusion, scoring code examples, and retrieval evaluation metrics.