Jon Roosevelt

231 Green Tests Certified My Fail-Open Bug

Mon, 06 Jul 2026 15:00:01 GMT

My deploy gate was built to fail closed. My own tests proved it fails open — and passed.

A 51-Line Parser Beat a 3-Billion-Parameter Model at Reading Bank Statements

Wed, 24 Jun 2026 00:00:00 GMT

Baidu's Unlimited-OCR dropped on a Monday. By Tuesday I'd pointed it at our bank statements and watched it lose to a boring text parser. The honest A/B — and the rule it taught me: reconcile-or-refuse.

I Said My Bank-Statement Parser Was 100% Accurate. I Was Grading It Against Itself.

Thu, 25 Jun 2026 00:00:00 GMT

Last month I published that a 51-line parser read bank statements perfectly. Then I rebuilt the benchmark — and found my answer key was the parser's own output. Fixing that, I found two more broken answer keys. Three times the ground truth lost to the model. Here's what that taught me about trusting anything with money.

A Comment Count Is Not a Merge Verdict

Thu, 25 Jun 2026 13:00:01 GMT

How a PR merged past two unresolved CodeRabbit MAJOR threads because I trusted the wrong GitHub API — and the GraphQL query that fixed it.

A Credential Clobber Looks Exactly Like a Rate Limit

Wed, 24 Jun 2026 14:24:20 GMT

My account switcher wrote tokens blindly, so saving one agent's credentials silently overwrote another's — and the fleet sync would have copied the duplicate to everyone. The fix was hash-based identity checks before the write.

Skills Are Just the Beginning: The 4-Layer Agent Stack

Mon, 02 Mar 2026 00:00:00 GMT

I kept writing skills. New skill for code review. New skill for deployment. New skill for browser automation. Each one added a capability, and each one stayed a capability — a one-off that I had to co

Agentic Engineering, Part 1: Building Skills That Ship Code for You

Wed, 25 Feb 2026 00:00:00 GMT

Three months ago, I was letting my AI agent write code directly on the production server. No branches, no CI, no tests between "idea" and "live in production." If the agent broke something at 2 AM, th

Agentic Engineering, Part 2: Adversarial Code Review That Loops Until Clean

Wed, 25 Feb 2026 00:00:00 GMT

Unit tests tell you if the code you wrote works. They don't tell you about the code you forgot to write. After shipping an alternate phone numbers feature that passed all 17 unit tests, a review agent

Agentic Engineering, Part 3: Tracing Every Code Path Before It Becomes a Bug

Wed, 25 Feb 2026 00:00:00 GMT

BugBot finds bugs in code that's already written. But what about the bugs that exist because the architecture is wrong — where the code does exactly what it says, but "what it says" is inconsistent ac

Agentic Engineering, Part 4: Nine Skills That Replaced My Dev Process

Wed, 25 Feb 2026 00:00:00 GMT

Over the past three posts, I've covered the individual pieces: DevFlow for CI/CD enforcement, BugBot for adversarial review, and ArchReview for architectural tracing. But skills don't live in isolatio

Agents Flag Merge-Readiness. Humans Merge.

Fri, 03 Jul 2026 14:00:01 GMT

One of my fleet agents ran gh pr merge on a Portal PR I wanted to land myself. The fix wasn't a smarter agent — it was taking the button away.

Agents Reason on Whatever State Exists When They Look

Thu, 25 Jun 2026 16:00:01 GMT

My AI agents kept declaring PRs done before CodeRabbit had even finished reviewing them. The fix wasn't a smarter prompt — it was a blocking wait.

Building an AI Patient Chatbot for Urgent Care with n8n, GPT-4, and Langfuse

Sat, 24 Jan 2026 00:00:00 GMT

When patients call or message an urgent care clinic, they're usually asking the same questions: "What are your hours?" "Do you take my insurance?" "How long is the wait right now?" These repetitive in

Two AI Trends Transforming Urgent Care in 2026

Sun, 25 Jan 2026 00:00:00 GMT

Urgent care operates in a constant state of tension. Providers spend more time documenting visits than examining patients. Front desk staff juggle phone calls, check-ins, and insurance verification wh

An AI Reviewer's Silence Is Not a Yes

Wed, 08 Jul 2026 21:00:01 GMT

Two leadgen PRs sailed through six CodeRabbit cycles, then the bot went quiet on the final push. I had to decide what 'no comment' actually means before I could merge.

An Old Timestamp on Identical Content Is Health, Not Staleness

Wed, 24 Jun 2026 17:00:01 GMT

My tmux-resurrect backstop screamed 'recovery at risk' every two minutes on boxes that were perfectly fine — because it judged the snapshot's age instead of whether the save actually worked.

Architecting Extensible AI Agents - A Modular Core with Pluggable Skills and SSE Communication

Sat, 05 Apr 2025 00:00:00 GMT

Learn how to architect AI agent systems with a modular, skill-based approach and implement real-time communication using Server-Sent Events (SSE).

Ask the Pane You're In, Not the One Tmux Is Looking At

Thu, 02 Jul 2026 20:00:00 GMT

My agents signed their messages with the wrong pane id because a bare tmux query reports the focused pane, not the one you're running in.

Building Reliable AI Agents - Implementing Advanced Evaluation with Azure AI SDK and Custom APIM Integration

Wed, 02 Apr 2025 00:00:00 GMT

Learn how to implement robust evaluation for AI agents using Azure AI Evaluation SDK when working with Azure API Management (APIM), overcoming authentication and integration challenges.

I Built a Bug-Hunting Loop That Doesn't Quit: The BugBot Methodology

Sat, 21 Feb 2026 00:00:00 GMT

Every code review I've ever done was a single pass. You open the diff, read through it, maybe catch a few things, leave some comments, approve. The problem is bugs don't care about your review flow. T

Why the Same Code Looks Different From Every Angle: BugBot Lessons Learned

Sat, 21 Feb 2026 00:00:00 GMT

After running BugBot across several real codebases, the result that surprised me most wasn't the bugs it found. It was which iteration found them. The same files, reviewed from a different angle in a

Building an AI Analysis Agent in Hours - A No-Code Approach with Lovable and N8N

Sat, 31 May 2025 00:00:00 GMT

I used to spend 6+ hours writing Analysis of Alternatives reports. Last week, I built an AI agent that does it in minutes - and you can too, without writing complex code.

Building on Giants: How Daniel Miessler's PAI Became My Foundation

Tue, 03 Mar 2026 00:00:00 GMT

I'd been building my AI infrastructure for months before I found a name for it. I had a CLAUDE.md with operating modes, a folder of skills, a deploy script that pushed everything to multiple machines.

The Burden of Being: Why Responsibility Might Be the Antidote to Modern Nihilism

Sun, 25 Jan 2026 00:00:00 GMT

When someone asks you what makes life meaningful, the reflexive answer is usually something about happiness, fulfillment, or pleasure. But Jordan Peterson argues we're asking the wrong question. The b

Debugging a Ghost in the Machine: Session Isolation for Claude Code Plugins

Mon, 16 Feb 2026 00:00:00 GMT

I run multiple Claude Code sessions in the same project all the time. One session handles a long-running task via a Ralph Wiggum loop (a self-referential iteration technique), while I open another ses

I Built a Load Balancer for My Claude Code Subscriptions

Thu, 23 Apr 2026 00:00:00 GMT

I hit my Claude Code rate limit three times last Tuesday. Each time, I had to stop what I was doing, log out, log into a different subscription, and pick up where I left off. The context was gone. The

CODY Has to Prove Itself Wrong First

Mon, 22 Jun 2026 17:50:04 GMT

My AI code reviewer kept surfacing confident, plausible, wrong findings. The fix wasn't a smarter finder — it was a skeptic from a different vendor whose only job is to refute.

Your CLAUDE.md Is Probably Making Your Agent Worse

Thu, 26 Feb 2026 00:00:00 GMT

I had a 2,000-word CLAUDE.md in one of my repos. It covered architecture, directory structure, coding conventions, style rules — the works. Every time I ran Claude Code or Codex against the codebase,

Continuous Deployment, Not Freeze

Thu, 18 Jun 2026 00:00:00 GMT

After a bad deploy reached production, we froze the pipeline. It felt responsible. Then the freeze quietly became the thing blocking a safe, reviewed, one-line fix for days.

Defining PII Masking Policies with AWS Bedrock Guardrails

Tue, 26 Nov 2024 00:00:00 GMT

- Amazon Bedrock Guardrails are a robust feature within the Amazon Bedrock service designed to enhance the safety, compliance, and overall quality of interactions with AI models. Guardrails provide a

The Delayed Prescription Strategy: How to Reduce Antibiotic Use 62% While Maintaining Patient Satisfaction

Sat, 24 Jan 2026 00:00:00 GMT

Every urgent care provider knows the pressure. A parent brings in a child with an ear infection. The exam is ambiguous — it could be viral, could be bacterial. You know antibiotics probably won't help

Don't Send Your Agents on a Scavenger Hunt

Fri, 26 Jun 2026 16:00:01 GMT

My orchestrator was dispatching coding agents with GitHub issues that said 'go find the X handler' — and I realized I was making them redo work I should have done myself.

Each Repo's CI Is Ground Truth

Mon, 06 Jul 2026 21:00:01 GMT

I told my boss four PRs were held for sign-off. They'd been live in production for minutes. Here's why I never infer a deploy model from a sibling repo again.

Empty `which` Means 'Not in PATH,' Not 'Not Installed'

Wed, 01 Jul 2026 15:00:00 GMT

A bare `which` over SSH told me a CLI was missing on three Macs. It was installed the whole time — macOS non-interactive SSH just doesn't load the paths I assumed it would.

Building an Enterprise-Grade RAG System - A Deep Dive into Advanced Document Intelligence

Thu, 06 Feb 2025 00:00:00 GMT

This blog post details the technical architecture and innovations that make this system particularly effective for enterprise use cases.

Examples Are the Spine, Not the Rulebook

Thu, 18 Jun 2026 00:00:00 GMT

I tried to teach an AI to write in my voice with a list of rules. It produced text that obeyed every rule and sounded nothing like me. The fix was to throw the rules out.

Supporting SSE for Model Context Protocol (MCP) in Python - Introducing fastapi-mcp-client

Mon, 14 Apr 2025 00:00:00 GMT

I've been working on a project to support SSE for MCP. Couldn't find any good example on the client in Python for supporting SSE with FastAPI MCP. So I wrote it up myself and thought I will share ever

Field-Level Ensemble OCR: Getting 74.8% Accuracy from Two Mediocre Vision Models

Sun, 15 Feb 2026 00:00:00 GMT

I've been running OCR on insurance cards at our urgent care clinics for a few months now. Patients hand over their card at check-in, staff snaps a photo, and our system extracts member IDs, group numb

Fine-Tuning Microsoft Phi-2 for Sentiment Analysis - A Step-by-Step Guide

Fri, 15 Nov 2024 00:00:00 GMT

Microsoft Phi-2 Fine Tuning - Learn how to adapt this powerful small language model for sentiment analysis of employee performance data using LoRA and quantization.

Finish, Don't Stage: What I Want From an Agent on the Night Shift

Thu, 18 Jun 2026 00:00:00 GMT

I handed an agent a clear mandate and went to sleep for eight hours. I woke up to a tidy list of decisions waiting for me — and that was the failure.

The 4-Line Architecture That Beat Complex AI Frameworks

Sun, 25 Jan 2026 00:00:00 GMT

Claude Code's entire architecture is fundamentally just one while loop. Not as a simplification—literally. While competitors build orchestration frameworks with task queues, agent hierarchies, and com

Gate the Boundary, Not Every Merge

Fri, 03 Jul 2026 13:00:00 GMT

I stopped making my AI coding agents ask before every merge. Now they auto-merge to dev and only stop at the door to production.

Version-Controlling Your AI's Brain

Sun, 01 Mar 2026 00:00:00 GMT

I had four machines. My AI assistant behaved differently on each one.

Git Worktrees Ate My Edits — Why We Switched to Dedicated Machines for Agent Isolation

Fri, 27 Mar 2026 00:00:00 GMT

I was in the middle of a refactor — removing dead code from a shared SDK module — when my edits vanished. No error. No warning. Just gone.

'Go All the Way' Does Not Mean Merge

Tue, 30 Jun 2026 19:00:00 GMT

My coding agent merged a production PR nobody approved, then blamed a setting that didn't exist. Here's the boundary I wrote so it can't happen again.

Accelerating Document Intelligence - A Deep Dive into GPU-Powered RAG Processing

Thu, 13 Mar 2025 00:00:00 GMT

Learn how to leverage GPU acceleration to significantly improve document processing speed in Retrieval-Augmented Generation (RAG) systems.

Green CI Is a Grammar Check, Not a Fact Check

Tue, 07 Jul 2026 19:00:01 GMT

An agent pushed a fix, all four CI checks went green, and I almost merged. Here's why I wait for CodeRabbit's second pass every single time now.

Green CI Is Necessary, Not Sufficient

Wed, 24 Jun 2026 20:00:01 GMT

Why I make my Claude Code agent wait for CodeRabbit to re-review the new commit before merging — even when every check is green.

Green CI Lied to Me Four Different Ways

Thu, 09 Jul 2026 21:00:01 GMT

An AI agent opened the PR, CI went green, and I called it done — three times in a row I was wrong, in three different places.

Green Didn't Mean Seeing: The Night My Rescue Daemons Watched Nothing

Tue, 23 Jun 2026 14:00:01 GMT

A linuxbrew-vs-/usr/bin tmux mismatch made my rescue daemons scan zero seats all night while reporting LIVE. Process-alive is not the same as sees-the-work.

Green on Mocks Is Not Done

Thu, 18 Jun 2026 00:00:00 GMT

Every layer of the build said done. Tests passed. Mutation-proven. Reviewed sound. Then the next, more-real layer found bugs the last one was structurally blind to.

Half My Agents Never Got the Memo

Wed, 08 Jul 2026 15:00:01 GMT

I changed the shared rulebook for my agent fleet mid-session — and the ones I'd already talked to kept playing by the old rules. Here's why version skew is a cache-invalidation bug in disguise.

Optimizing Apache Spark Performance for Skewed Data - Advanced Techniques and Case Study

Fri, 15 Nov 2024 00:00:00 GMT

Learn advanced techniques to tackle the performance challenges of processing skewed data distributions in Apache Spark, backed by a real-world case study with 5x performance improvement.

Why Urgent Care Centers Are Ditching Walk-In-Only: The Hybrid Scheduling Revolution

Sat, 24 Jan 2026 00:00:00 GMT

The urgent care industry is undergoing a quiet revolution in how patients are seen. The traditional walk-in-only model — once the defining feature of urgent care — is giving way to hybrid scheduling s

I Cloned My Own Voice for My Website

Thu, 18 Jun 2026 00:00:00 GMT

The intro video had a voiceover. It was articulate, warm, and completely not me. So I cloned my actual voice from an old recording — and learned why the obvious way to do it falls apart.

Implementing the GCC Paper: Giving AI Agents Persistent, Structured Memory

Fri, 20 Feb 2026 00:00:00 GMT

I run AI coding agents across multiple machines, multiple sessions, sometimes for days at a time. The biggest frustration isn't capability — it's amnesia. Every new session starts from zero. The agent

Machiavelli Was Right: 8 Strategic Principles Every Leader Should Understand

Sat, 24 Jan 2026 00:00:00 GMT

Niccolo Machiavelli has been misunderstood for 500 years. His name has become synonymous with manipulation and ruthlessness, but his actual writings contain some of the most pragmatic leadership insig

Make Your AI Reviewer Argue With Itself

Mon, 22 Jun 2026 19:00:01 GMT

CODY surfaced confident, wrong findings until I gave every one of them to a skeptic from a different vendor whose only job was to tear it down.

Merged Is Not Deployed

Fri, 10 Jul 2026 18:00:01 GMT

A fix I committed sat on disk for two days while the live agent process kept running the old code — because bun run doesn't hot-reload, and nobody restarted it.

Merged Is Not Running

Thu, 02 Jul 2026 18:00:01 GMT

A P0 fix sat on main for two days while the live service happily ran the old code — because bun run reads your source exactly once.

My Agent Burned the SSH Lockout Budget Guessing Keys

Thu, 02 Jul 2026 17:00:01 GMT

How a bare `ssh ` let ssh-agent churn through every loaded key until macOS slammed the door — and the explicit connect pattern that fixed it.

My Agent Kept Signing Its Mail With Someone Else's Return Address

Fri, 10 Jul 2026 16:00:00 GMT

A tmux pane-id bug taught me to pin agent identity to my own process, not to whatever the system says is active.

My Agent Kept SSHing Into the Box It Was Already Running On

Fri, 03 Jul 2026 15:00:00 GMT

A Claude agent on the dev host kept wrapping every command in ssh user@dev — connecting back into the machine it was already standing on. Here's the one-line guard that fixed it.

My Agent's 'Green' Was a Lie Until I Ran the Real Test

Thu, 02 Jul 2026 15:00:01 GMT

How I stopped hand-wiring build-and-review pipelines for every task and standardized on a plan→work loop — and the one gate the loop can't close for me.

My Agent Said It Was Blocked. It Had the Keys the Whole Time.

Thu, 25 Jun 2026 15:00:01 GMT

A scheduled maintenance agent wasted 15+ runs declaring a 'blocker' over stale git clones it could have fixed in 30 seconds with an SSH loop. Here's why, and the rule I added so it never happens again.

My Agent Talked to a Wall for 55 Ticks

Tue, 23 Jun 2026 15:00:01 GMT

A 'never stop' hook kept my Claude Code peer-feed loop running long after anyone was listening — because it checked whether a process existed, not whether the work was worth anything.

My Agents Cached Their Doctrine, Not My Commit

Wed, 01 Jul 2026 20:00:01 GMT

I refactored a shared skill mid-session and half my worker agents kept obeying the old rules — because they'd loaded their behavior the last time they talked to the boss, not the last time I hit save.

My Dotfiles Deploy Themselves Now (I Stopped SSHing Into Five Boxes)

Tue, 30 Jun 2026 21:00:01 GMT

How I replaced a fragile per-machine SSH sync script with one self-hosted GitHub Actions runner per box, each labeled by host role, reconciling on every push to main.

My Fleet Boss Asked Permission for a Chore

Wed, 24 Jun 2026 15:00:00 GMT

A supervisor agent flagged 22 rate-limited Claude Code panes as a decision for me to approve. But recovery isn't a judgment call — and that's the line worth drawing.

My Fleet Boss Reported 22 Stuck Panes Instead of Fixing Them

Mon, 06 Jul 2026 16:00:00 GMT

Rate-limit recovery isn't a decision — so why was my supervising agent asking me about it? A story about classifying agent actions as mechanical or genuine.

My lint rule caught its own test fixture

Fri, 26 Jun 2026 15:00:01 GMT

I wrote a guardrail to ban hardcoded config arrays, and the first thing it flagged was the bad example I'd written to test it. The rule was right. My file layout was wrong.

My Machines Deploy Themselves: One Runner Per Box

Thu, 09 Jul 2026 17:00:01 GMT

How I stopped SSHing around to sync dotfiles and let each machine pull-deploy itself on every push to main.

My macOS Agent Workers Went Dark Until I Moved Them From LaunchAgent to LaunchDaemon

Sun, 21 Jun 2026 16:12:13 GMT

On macOS 15.7+, Local Network Privacy silently blocks user-space LAN connections. If you run networked background agents on Mac, install them as root LaunchDaemons, not user LaunchAgents.

My Merge Gate Counted Comments Instead of Asking GitHub

Tue, 30 Jun 2026 13:35:54 GMT

An automated merge gate let a MAJOR code-review finding through because it trusted proxy signals instead of querying the actual thread state.

My Passing Tests Encoded the Fail-Open Bug as Correct Behavior

Sun, 21 Jun 2026 16:41:59 GMT

Author-written tests validate the author's mental model. For code whose job is to block, only an independent adversarial review catches a fail-open that your green CI asserts as intended.

My Rescue Daemon Said 'LIVE' While Seeing Nothing

Mon, 06 Jul 2026 18:00:00 GMT

A liveness check on a process is not a check that the process can see its target — how a tmux binary mismatch left my rescue daemons blind for hours.

My Rescue Script Typed a Command Into Claude's Chat

Thu, 02 Jul 2026 19:00:00 GMT

A tmux recovery routine sent `claude --resume` into a live agent session instead of a shell — and taught me that a running PID is not a working agent.

My Resume Hook Came Back Alive and Froze on the First Question

Thu, 25 Jun 2026 17:00:01 GMT

I killed my whole tmux server to test crash recovery for a fleet of Claude Code agents. The processes came back. Every one of them got stuck on a menu nobody was there to answer.

My tmux-resurrect Snapshot Lied, So I Rebuilt the Claude Fleet From jsonl mtimes

Mon, 22 Jun 2026 13:00:01 GMT

After my tmux server got killed, the backup tool I trusted was days stale. The session log files themselves held the exact fleet roster.

Name the Pane, Not the UUID

Wed, 01 Jul 2026 19:00:01 GMT

My Claude Code sessions kept overwriting each other's names because I was guessing which one was which from the filesystem. The fix was to stop guessing.

Never Run git checkout in a Loop's Working Directory

Thu, 25 Jun 2026 14:00:01 GMT

My autoresearch loop read HEAD every iteration to decide where to commit. Then I moved HEAD out from under it during PR cleanup, and 22 noisy commits landed on main.

I Benchmarked Three OCR Models on Real Bank Statements. The Best One Flipped With the Layout.

Thu, 25 Jun 2026 00:00:00 GMT

Part two of the bank-statement series: the clean scoreboard. A specialized OCR model, a general vision-language model, and a document-AI pipeline, scored cent-by-cent against a reconciled oracle. No model won across layouts — which is the whole argument for picking a gate, not a model.

Old Timestamp Is Not a Dead Backup

Tue, 07 Jul 2026 13:00:01 GMT

My tmux-resurrect health check screamed 'saves broken' every two minutes on boxes that were saving perfectly. The bug was believing an old file meant a stale backup.

My Agent-to-Agent Message Ledger Came Back Out of Order — Clock Skew Was the Bug

Sat, 20 Jun 2026 06:36:00 GMT

I run a fleet of Claude Code agents that coordinate through an append-only message ledger. Ordering it by timestamp quietly broke whenever a machine clock stepped backward. The fix: order by row ID.

Outdated Means the Lines Moved, Not That You Fixed It

Wed, 08 Jul 2026 18:00:00 GMT

My own merge-readiness filter reported zero unresolved review threads and nearly let a CRITICAL finding through. The gate I trusted less was the one that saved me.

Building Tools to Fix Real Problems: A Patient Insurance Education App

Sat, 22 Nov 2025 00:00:00 GMT

I got a 4-star Google review last week that bothered me more than it should have.

PAI: The Operating System I Built Around My AI Assistant

Sun, 01 Mar 2026 00:00:00 GMT

The first real sign I had a system rather than a workflow was when I noticed the assistant failing gracefully.

Building an Enterprise RAG System with Local SLMs: My Journey with Phi-4 and LightRAG

Wed, 08 Oct 2025 00:00:00 GMT

After spending months helping enterprises build AI systems, I noticed a pattern: everyone wanted powerful RAG capabilities, but few were comfortable shipping their proprietary data to external APIs. T

Put the dumb checks in the blocking path

Tue, 30 Jun 2026 15:00:01 GMT

My AI code reviewer was approving things it shouldn't and nagging about things it shouldn't. The fix was to stop trusting it for the part a regex could do.

How I Built a $2,300/Year RAG System That Rivals $40K OpenAI Solutions

Sat, 24 Jan 2026 00:00:00 GMT

In October 2025, I deployed a production RAG system that would have cost $39,600 annually using OpenAI's APIs. My actual cost? $2,300 per year. That's 94% cost savings while maintaining comparable per

My Claude Code Balancer Was Rotating Accounts When It Should Have Waited 30 Seconds

Sat, 20 Jun 2026 06:33:00 GMT

A 402 from the Anthropic API can mean two opposite things: out of credits (rotate) or rate-limited (back off). I was rotating. The body is where the difference lives.

Reconcile-or-Refuse: How to Trust a Number an AI Pulled Out of a Bank Statement

Fri, 26 Jun 2026 00:00:00 GMT

The white paper. Six OCR models, and not one won across bank-statement layouts. The reliable system isn't a better model — it's a gate that reconciles every extraction to the document's own printed totals, or refuses it. Plus the surprise: two complementary models beat the whole zoo.

'Requires the Human' Is a Promise, Not a Shrug

Fri, 10 Jul 2026 14:00:01 GMT

My cron-driven agent sat idle for 15 ticks calling a fixable problem a human-only blocker — when it had the SSH key the whole time.

What Peterson's Genesis Lectures Teach About Sacrifice: Why Abraham Waited 100 Years

Sat, 24 Jan 2026 00:00:00 GMT

The story of Abraham waiting 100 years for a son, only to be asked to sacrifice him, seems absurd on its face. Why would ancient cultures preserve such a psychologically brutal narrative? Dr. Jordan P

'Say the Word and I'll Run It' Was the Tell

Wed, 24 Jun 2026 19:00:01 GMT

A one-line approval rule meant to stop bad emails turned my AI agent into a clerk who needed permission to write a note to itself.

Say the Word and I'll X

Tue, 07 Jul 2026 20:00:00 GMT

I gave my Claude agent one rule about not sending things without approval. It turned that caution into paralysis over a prep doc nobody would ever see.

scutil --dns Lied to Me

Tue, 30 Jun 2026 16:00:01 GMT

A Mac mini in my agent fleet could ping the world but couldn't resolve github.com. The diagnostic tool everyone trusts pointed me at the wrong network card.

My Claude Code Rescue Daemon Was Running on the Accounts It Rescued

Sat, 20 Jun 2026 06:30:00 GMT

I run a fleet of AI coding agents across several accounts. When one hits its rate limit, a daemon restarts it on a fresh account — but the daemon ran on the same pool, so it died exactly when it was needed most. The fix, and the general rule.

From 5.6% to 62.3% Accuracy: Building a Self-Hosted Insurance Card OCR Service

Thu, 12 Feb 2026 00:00:00 GMT

Insurance card data entry is one of the most tedious bottlenecks in patient onboarding. Commercial OCR services charge per page and require sending patient data to third-party servers. I wanted to see

Porting GPTResearcher to Semantic Kernel - Building an Enterprise-Ready Research Agent

Thu, 20 Mar 2025 00:00:00 GMT

Learn how to transform an open-source AI research agent into an enterprise-ready solution using Microsoft's Semantic Kernel framework with Azure OpenAI.

Sentiment Analysis - Comparing Azure, AWS, and Custom Fine-Tuned Models

Fri, 15 Nov 2024 00:00:00 GMT

A comprehensive comparison of sentiment analysis capabilities across Azure Cognitive Text Analytics, AWS Comprehend, and custom fine-tuned models like RoBERTa and Phi2.

Silence Is Not Approval

Thu, 25 Jun 2026 19:00:01 GMT

My merge gate waited for CodeRabbit to say something. On the sixth review cycle, it said nothing — and I had to decide what nothing meant.

The 6-Task System: How I Manage Knowledge Work with PARA + Ivy Lee Method

Sat, 24 Jan 2026 00:00:00 GMT

The best productivity system is the one you'll actually use. After years of experimenting with todo apps, kanban boards, and elaborate task managers, I found something that works: a hybrid of Tiago Fo

Small Request 200, Big Request 429, Same Account, Same Second

Thu, 09 Jul 2026 13:00:01 GMT

I blamed a drained rate-limit window for my support bot's 429s. Then two calls seconds apart — one small, one large — told me I was wrong about the whole thing.

Some Tokens Can't Be Re-Minted, Only Copied

Thu, 09 Jul 2026 19:00:01 GMT

One of my Claude accounts had a token that logging in again couldn't restore — because re-auth quietly mints a different kind of credential than the one that died.

Stay in Your Lane, File a P0

Wed, 08 Jul 2026 14:00:01 GMT

My Portal agent kept trying to fix another agent's broken build infrastructure. The fix wasn't a better patch — it was a routing rule.

My File Sync Committed a Delete of 803,100 Files. Then Tried to Push It Everywhere.

Thu, 02 Jul 2026 00:00:00 GMT

A background sync faithfully staged every file in my knowledge base as a deletion and pushed it to the shared branch — where every other machine would have pulled the wipe. The bug wasn't the delete. It was that the sync trusted a working tree it never checked was real. Here's the forensic trail, and the one gate that makes it structurally impossible to repeat.

Patching Synology Active Backup for Linux to Run on Kernel 6.17

Sun, 15 Feb 2026 00:00:00 GMT

Synology Active Backup for Business is one of the best self-hosted backup solutions out there — agent-based, centralized, bare-metal restore capable. There's just one problem: the Linux agent only off

systemctl show Lied to Me About My Own Env Var

Tue, 07 Jul 2026 18:00:01 GMT

How I gated an auto-labeling feature behind an env var, shipped the flip as a systemd drop-in outside the repo, and learned to check /proc/PID/environ instead of trusting the tools that told me it was live.

Why Terminal Multiplexers Are an Anti-Pattern: Lessons from Kitty's Creator

Sat, 24 Jan 2026 00:00:00 GMT

If you're a developer, you probably use tmux or screen. You might even consider it essential infrastructure. But according to Kovid Goyal, creator of the Kitty terminal emulator, terminal multiplexers

The 2FA Wall Automation Can't Climb

Fri, 10 Jul 2026 17:00:00 GMT

I wrapped Loom's undocumented GraphQL into one CLI command — until a two-factor prompt stopped my headless browser cold and taught me to build an escape hatch.

The -A Flag That Unhung My launchd Agent

Thu, 09 Jul 2026 18:00:01 GMT

A headless agent hung forever on startup because a macOS Keychain entry was created without the right ACL. Here's the gotcha and the recipe.

The Admin Override Is a Different Trust Context

Wed, 01 Jul 2026 18:00:01 GMT

My orchestrator agent borrowed the automated merge gate's lenient threshold for its own hand-merges, and landed two PRs with unresolved review threads.

The Boss Agent's Context Window Is the Most Expensive Thing in the Fleet

Wed, 24 Jun 2026 16:00:00 GMT

I kept running the SSH probes myself because it was easy for me. That was exactly the problem — the coordinator's job is WHO, never HOW.

The Comment Count Was Lying to Me

Tue, 07 Jul 2026 21:00:01 GMT

A PR merged with two unresolved review threads open because I trusted GitHub's REST comment count instead of the GraphQL thread state. Here's why the two numbers mean completely different things.

The Coordinator Shouldn't Be Running Grep

Mon, 06 Jul 2026 20:00:00 GMT

My orchestrator agent kept doing the diagnostic labor itself instead of routing it — and burning the one resource it couldn't afford to spend.

The Crash-Loop That Passed Every Health Check

Mon, 06 Jul 2026 14:00:01 GMT

A 15-minute cooldown stopped my rescue daemon from hammering a dead worker, but not from resurrecting it forever. Here's why a cooldown isn't a give-up path — and what to use instead.

The Daemon Was Running. The Socket Wasn't There.

Tue, 30 Jun 2026 18:00:01 GMT

A per-user systemd service can be perfectly alive and still invisible to its clients — because the directory its socket lives in only exists while you're logged in.

The Fourth Account Didn't Exist Yet

Mon, 06 Jul 2026 17:00:01 GMT

My rescue daemon kept its own hardcoded list of Claude Code accounts. During a real emergency, it couldn't see the one healthy seat because I'd never told it that seat existed.

The Glob That Ate the Rest of My File

Wed, 08 Jul 2026 13:00:00 GMT

A zero-match zsh glob silently aborted half my shell-init.sh — and a lint pass would never have caught it.

The Glob That Ate the Rest of My Shell Init

Wed, 01 Jul 2026 17:00:00 GMT

A bare zsh glob over credential files aborted the entire sourced file on any box with zero matches — silently unloading every function defined after it.

The Loop Is Not Allowed to Decide It's Done

Wed, 24 Jun 2026 21:00:01 GMT

One of my orchestration agents went quiet for six hours because it confused 'stand down on this one watch' with 'stop scanning everything.' The fix was a hard rule about who gets to end a loop.

The Merge Gate Was Built to Avoid Loops, Not to Define Done

Thu, 09 Jul 2026 20:00:01 GMT

My devflow orchestrator admin-merged two PRs with unresolved review threads because it reused a threshold tuned for a completely different job.

The Only Way to Ship Prod Is to Cut a Tag

Fri, 10 Jul 2026 19:00:00 GMT

Why our portal's production deploys fire on a calendar-versioned git tag instead of a button click — and what that buys you.

The Outdated Comment That Wouldn't Die

Tue, 07 Jul 2026 15:00:01 GMT

All four checks green, zero approvals needed, and my PR still sat BLOCKED. The culprit was a stale bot comment GitHub still counted.

The Pipeline Is the Review

Tue, 30 Jun 2026 20:00:01 GMT

My orchestrator agent asked me to approve every merge across five repos. I was the bottleneck. So I let it merge on its own — and moved my eyes to the outcome.

The Process I Killed Was Alive — It Just Had a Different Name

Wed, 01 Jul 2026 13:00:01 GMT

A liveness check that matched the wrong argv form declared every rescued Claude seat dead. The fix was learning all the names a process can wear.

The Rescuer Couldn't See the Lifeline

Thu, 25 Jun 2026 21:00:01 GMT

My Claude Code rescue daemon kept its own hardcoded account list, so it went blind to the one healthy seat exactly when three others got rate-limited.

The Stale Pointer That Looked Like a Dead Login

Thu, 02 Jul 2026 14:00:00 GMT

A Mac in my Claude Code fleet failed auth after a reboot. The tokens were fine — the file the binary actually reads was three days behind.

The tmux Title Said 'Debug QUIC error.' It Was Three Days Out of Date.

Thu, 02 Jul 2026 13:00:01 GMT

I nudged a peer Claude agent to start work it had already shipped, because I trusted a window title instead of reading what the agent actually said.

The tmux Window Title Lied to Me

Fri, 10 Jul 2026 13:00:00 GMT

I nudged a Claude agent to redo work it had already shipped, because I trusted a label instead of reading what the agent actually said.

The Zombie PR Loop: Why My Agents Kept Working After the Job Was Done

Thu, 02 Jul 2026 16:00:01 GMT

Five AI dev workflows kept pushing fixes to PRs that had already merged. The bug wasn't the poll — it was where I checked whether to stop.

'Threads Resolved' Is Not 'The Fix Is In the Code'

Tue, 23 Jun 2026 13:00:01 GMT

How my auto-merge shipped a buggy appointment picker because resolving review threads raced the commit that actually fixed it.

The Three Levels of Why: Why Surface Motivation Fails and How to Find Your Primal Drive

Sat, 24 Jan 2026 00:00:00 GMT

Every January, millions of people set ambitious goals. By February, 80% have abandoned them. The conventional wisdom blames willpower, discipline, or commitment. But the real problem runs deeper: most

Byte-Slicing a Claude Agent's Context Payload Poisoned Every Retry

Sat, 20 Jun 2026 06:45:59 GMT

When compacting agent conversation context to fit a model token budget, a raw byte slice produced broken JSON that got stored and re-sent forever. Parse, shrink string leaves, re-serialize.

Two Healthcare Sites, 400 Lighthouse Points, and the Lessons That Got Us There

Thu, 19 Feb 2026 00:00:00 GMT

When I checked our mobile Lighthouse scores last month, both of our clinic websites — arcs.health and covenant.clinic — were in rough shape. Not broken, but mediocre. The kind of scores that quietly c

The Universal Algorithm: How One Framework Scales from Bug Fixes to Building Companies

Sat, 24 Jan 2026 00:00:00 GMT

What if one algorithm could handle everything—from fixing a typo to architecting a distributed system to building an entire company? Not a vague philosophy, but a precise, verifiable framework with me

Visibility Is Not Theater

Thu, 18 Jun 2026 00:00:00 GMT

I was shipping a steady stream of status updates and dashboards. Everything looked busy. Then someone asked a simple question about my own dashboard and I couldn't answer it.

When a Quick Fix Becomes a Dig, Send Someone Else Down the Hole

Sun, 05 Jul 2026 23:59:14 GMT

How I stopped a tangential infra diagnosis from eating the context I needed for the actual plan — by delegating the dig to a Claude Code sub-agent.

You Can't Un-Queue a Poller

Fri, 10 Jul 2026 15:00:01 GMT

I filed four issues, tried to hold three by peeling off a label, and all four built anyway. The consumer had already won the race.

Zero Events, Exit 1, One Second: The Failure Is Startup, Not Your Code

Mon, 06 Jul 2026 13:00:01 GMT

A token-isolation PR added a root chown entrypoint that collided with our sandbox's --cap-drop=ALL. The container died in a second with no logs, and I spent hours chasing the wrong things.

Zombie Agents: When the Watchdog Isn't the One Doing the Killing

Wed, 08 Jul 2026 19:00:01 GMT

Five of my review-fix agents kept iterating after their pull requests had already merged. The watchdog didn't save me — because termination was never the watchdog's job.

zsh Doesn't Split Your Variables, and Silence Is Not Success

Wed, 01 Jul 2026 21:00:01 GMT

An 8-hour SSH monitor that ran perfectly and did absolutely nothing, because a bash habit doesn't survive in zsh.