Writing

2026

• When Consciousness Fully Stops, How Does the Mind Return?Oct 18Substack
• Can You Tell an Eval from a Review?Oct 13Dev.to
• Whoa… My Claude Code Skill Descriptions Are Way Too Long…?!Oct 9Dev.to
• Is a README for Humans or for LLMs?Oct 9Dev.to
• Turning Claude Code into a Writing Harness with Claude Mods and an Output StyleOct 3Dev.to
• "Is There Any Point to This?" Removing the Jev Plugins I Added to Claude Code After One WeekSep 29Dev.to
• Why Does AI Writing Feel Like It's Floating?Sep 28Substack
• Before Adding a Jev-Type Judgment Model, Read Your Generation Model's ProbabilitiesSep 27Dev.to
• Moving My Research Pipeline's Judgment Calls from an LLM to Jev, a Judgment-Only ModelSep 25Dev.to
• What Does It Take to Reproduce Jev's Decisions Locally?Sep 24Dev.to
• I Added Jev's Skill Router to Claude Code and Turned Back Just Before Rewriting the Skill ListingSep 21Dev.to
• How Close to Opus Does Jev, a Model That Writes No Text, Get at Skill Selection in 0.3 Seconds?Sep 21Dev.to
• I Made the Top Model My Session Default, and One Heavy Implementation Plus Its Review Burned Through Fable's Usage LimitSep 17Dev.to
• People Inside the Business Don't Call It a 'Domain'Sep 16Dev.to
• Why an 11-Second Burst Never Showed Up in the Log That Gets Read Every WeekSep 14Dev.to
• I Deleted the LLM-Facing Architecture Docs I Had Committed 159 Times in 3 Months: Structure Goes to LSP, Reasons to ADRs, Diagrams to HumansSep 5Dev.to
• I Asked AI to Diagnose My Knowledge Blind Spots — 15 Days of Deadlock Moved in 85 MinutesSep 2Dev.to
• My Dead-Code Scan Returned Zero, Then I Deleted 2,063 Lines: Detectors Measure References, Not ConsumptionAug 31Dev.to
• After Cutting My AI Reviews, I Put a Complexity Ceiling in RuffAug 30Dev.to
• I Cut My AI Review Chain From 6 Stages to 1: Breaking the Loop That Never Hits Zero FindingsAug 29Dev.to
• My KPIs Improved. I Deleted My Homegrown LLM Judge AnywayAug 27Dev.to
• Two Drafts Passed Every Eval and Both Were Hollow. Attach the Raw Transcript to the Ledger You Hand Your AIAug 21Dev.to
• I Handed 41 Tasks to an AI Loop. The Bottleneck Was Judgment, Not CodeAug 20Dev.to
• AI Review Kept Creating Work: Why I Deleted 4,541 LinesAug 16Dev.to
• Can Six-Month-Old AI Code Survive Today's Review? A 25-Bug TriageAug 14Dev.to
• After I Ran Out of Things to Ask AI, One Desire RemainedAug 13Substack
• Claude Code's Memory Has No Vectors — Try ADRs Before Memory RAGAug 6Dev.to
• On Embodiment in the Age of AIAug 4Substack
• Give Claude Code a Second HarnessAug 3Dev.to
• Banned from Wikidata Overnight — I Believed Every Edit Was Compliant, but All 109 Items Were Deleted as PromotionJul 30Dev.to
• What Humans Should Approve Is Intent, Not the Diff — A Decision Table for Agent Approval GatesJul 29Dev.to
• claude-security Beta: Cost, Output, Accuracy — 189 Agents, 2 Hours, MeasuredJul 28Dev.to
• Opus 5 Changed How Rules Should Be Written — Audit YoursJul 27Dev.to
• herdr, a tmux for AI Agents — Until the Editor DisappearedJul 21Dev.to
• What Do My AI Agent's Logs Look Like in OpenTelemetry?Jul 17Dev.to
• Fault Injection TDD Found 3 Silent Failures in My LLM AgentJul 16Dev.to
• LLM-as-Judge Shouldn't Aggregate Scores: Binary Checks as Evidence, One Holistic VerdictJul 15Dev.to
• Building an Autonomous Agent on an M1 Mac, by ChoiceJul 14Dev.to
• Why Did My Agent Decide That? 3 Observability PatternsJul 13Dev.to
• Claude Code from iPhone: Plugging 3 Holes in Remote ControlJul 9Dev.to
• Linters Can't Measure a Non-Code Asset's Value — I Built an LLM Stocktake for ItJul 8Dev.to
• I Built a Skill for Easy Codex Reviews from Claude CodeJul 7Dev.to
• You Got It Right — But You Didn't Know ItJul 2Substack
• When Consciousness Fully Stops, How Does the Mind Return?Jun 14Substack
• Where Does the Accountability Externalized by AI Go?May 24Substack
• Is ReAct Needed in Production? — Separating Design and Operation PhasesMay 1Dev.to
• (3) The LLM Workflow Quadrant Is Missing from Our VocabularyApr 30Dev.to
• Where ReAct Agents Are Actually Needed in BusinessApr 29Dev.to
• Organic Growth and Content Integrity in an AI Writing TeamApr 18Dev.to
• Building Zed as an Observation Window for Claude Code — Japanese Typography with IBM PlexApr 16Dev.to
• AI Agent Black Boxes Have Two Layers — Technical Limits and Business IncentivesApr 14Dev.to
• Can You Trace the Cause After an Incident?Apr 13Dev.to
• Freedom and Constraints of Autonomous Agents — Self-Modification, Trust Boundaries, and Emergent GameplayApr 1Dev.to
• Porting Game Dev Memory Management to AI Agent Memory DistillationMar 30Dev.to
• Where to Put a Coding Agent's Knowledge — and How to Make It StickMar 26Dev.to
• The Install and Hope Problem with MCP ToolsMar 24Dev.to
• My Agent's Memory Broke — A Day Wrestling a 9B ModelMar 23Dev.to
• Not Reasoning, Not Tools — What If the Essence of AI Agents Is Memory?Mar 21Dev.to
• Natural Language as Architecture — Controlling an Autonomous Agent with Prompts, Memory, and Fail-Safe DesignMar 14Dev.to
• Prompt-Based Alignment Has a Ceiling — 3-Model Prisoner's Dilemma EvidenceMar 11Dev.to
• I Built an AI Agent from Scratch Because Frameworks Are the VulnerabilityMar 8Dev.to
• I Made a Paper I Don't Understand Run in the Browser — Active Inference × Claude CodeMar 7Dev.to
• Obsidian's Official CLI Is Here — No More Hacking Your Vault from the Back DoorMar 6Dev.to
• Stop Using Default Settings — 10 Claude Code Configs That Actually WorkMar 5Dev.to
• I Found the Shortest Route to Ship My Claude Code Skills to 57,000 DevelopersMar 4Dev.to
• I Gave Claude Code a Simulator and It Started Tapping and Taking Screenshots on Its OwnMar 3Dev.to
• Kimi Wrote 8,500 Lines, Blamed Me for Delegating, Then Claimed to Be ClaudeFeb 28Dev.to
• Embedding Memory into Claude Code: From Session Loss to Persistent ContextFeb 27Dev.to
• Kimi Killed 4 of Claude's Best Ideas — An AI Peer Review in PracticeFeb 26Dev.to
• I Had Gemini Explain Its Own Deep Research — Half Was Self-JustificationFeb 25Dev.to
• Offloading AI's Weak Spots to Shell Scripts — Designing, Building, and Publishing a Skill Audit CommandFeb 24Dev.to
• Building a Claude Code × Kimi K2.5 Hybrid Environment: Opus Designs, K2.5 ImplementsFeb 24Dev.to
• The Real Power of Claude Code Isn't Code Generation — It's Autonomous Context OrchestrationFeb 23Dev.to
• Before and After Alignment — I Typed 'Hello' Into a Base Model and Got an Anime ReviewFeb 22Dev.to
• What I Learned from Breaking My AI Pipeline for Two Days StraightFeb 22Dev.to
• I Tried an Opus Orchestrator and Killed It — The ROI of Multi-Agent SystemsFeb 21Dev.to
• Zero Python Code: How I Built a Daily AI Research Report SystemFeb 15Dev.to
• 15 Days of Skill Sprawl in Claude Code — Lessons from 3 AuditsFeb 14Dev.to
• Running Claude Code from iPhone via SSH + tmuxFeb 14Dev.to
• Cursor to Zed: Disabling Built-in AI for a CLI-First SetupFeb 12Dev.to
• The Day I Told Claude Code 'You ARE the LLM'Feb 11Dev.to
• Never Trust LLM Output — 6 Defenses from Building a PDF-to-Anki CLIFeb 11Dev.to
• A Beginner's First 10 Days of Real Development with ECCFeb 9Dev.to