Blog

Practical writing on AI systems, knowledge infrastructure, and the mechanics of measurable delivery.

Latest dispatch

Blog Publishing Pipeline Jun 15, 2026 10 min read

Building a Multi-Channel Blog Publishing Pipeline with AI

For a while, publishing a single blog post was a small act of bookkeeping I dreaded. Write the piece, paste it into the site, then re-format it for the newsletter, then re-format it again for syndication, then chase down the canonical link so I wasn't quietly competing with myself in search resul...

Blog Publishing Pipeline Ai Assisted Writing

Read the article

How to Prep for a 6-Session On-Site Interview

Six sessions. Seven interviewers. One day. Each interviewer comes in with a different lens...

A technical hourglass filters source notes through a verification check into a clean Weibull model, with six day markers leading to a prepared outcome.

Ai Jun 6, 2026 10 min read

How AI Compresses Technical Prep: A Weibull Refresher Case Study

Six days, six interview sessions, seven interviewers, and a reliability-engineering topic I hadn't taught in years. Here's the AI-assisted prep system that turned a panic window into focused, …

Security May 21, 2026 7 min read

I Let AI Audit My Own Site — And Found 12 Real Vulnerabilities

I asked an AI to break into my own production Django site. It found twelve real vulnerabilities — two critical. Here's the full list, what the AI got …

A single task splitting into three phases — rough sketch strokes for exploration, geometric planning diagram, polished ink-on-cream for synthesis — flowing from one input to one output

Cognitive Routing May 19, 2026 15 min read

Hybrid Runs: Cheap Exploration, Expensive Polish

A task is not one thing. It's exploration, planning, and synthesis — each with different model requirements. Here's how to break the monolithic model assumption and route per-phase …

A workhorse model handling routine tasks at center, with a premium model standing by for escalation — the economic default flipped

Cognitive Routing May 16, 2026 21 min read

The GLM Default: Making Model Economics Work

Default to the cheap model and justify escalation to the expensive one. Here's why flipping the burden of proof produces better AI systems — and better economics.

Two separate paths diverging from a central AI core — one cognition depth, one model selection

Cognitive Routing May 11, 2026 8 min read

Why Cognitive Routing and Model Routing Are Different Problems

Most AI assistant architectures conflate two decisions: how hard should I think? and how much should I spend? Here's why they need to be separate control layers.

A metric pedestal on a cracked foundation — local optimization undermining system health.

Process Improvement May 9, 2026 6 min read

When Optimization Backfires: The Psychology of Departmental Metrics

When departments optimize their own metrics, the system degrades. The mechanism isn't mathematical — it's psychological. Why metric ownership becomes identity, gaming is rational, and the fixes are …

Hand mirror reflecting a network of connected nodes

AI Practice May 7, 2026 6 min read

Co-Learning with Agentic Frameworks: Building Your Cognitive Augmentation Without Losing Your Mind

When you build a custom AI assistant that knows your patterns and shapes how you think, you can lose track of what's real. Here's how to co-learn with …

Looking glass zoomed on a graph network, representing supervised agency and structured human-AI collaboration.

AI Practice Apr 27, 2026 6 min read

Supervised Agency: A Day in the Life of Human-AI Collaboration

How a 37-point Lighthouse improvement revealed the patterns that make human-AI partnerships work.

A support answer becomes a reviewable scaffold bridge that reaches an executing JSL workflow and analytical chart.

JMP Apr 20, 2026 5 min read

From Support Answers to JSL Workflows

The JMP MCP Expert System started by answering questions. But the real request was never just an answer—it was executable workflows. This is the story of how session-to-JSL …

Fragmented JMP sources cross visible MCP checkpoints to an analytical workbench with an inspectable provenance path.

Support Architecture Apr 20, 2026 6 min read

Building a Source-Backed JMP Expert System with MCP

This is not just a chatbot story. It is a story about turning fragmented JMP support knowledge into a source-backed assistant surface with visible tool boundaries, provenance, and …

A magnifying lens over source documents, revealing verified details — source-backed expertise in action.

Support UX Mar 27, 2026 4 min read

Why AI Support Tools Fail When Users Cannot See the Evidence

Fluent answers are not enough. If a user cannot inspect the evidence path behind an AI support answer, the product is optimizing for persuasion before trust.

A public telemetry dashboard and mobile system view showing live status, traces, and visits.

Working in Public Mar 16, 2026 4 min read

The New Transparency

I built a dashboard for my personal site. Then I made it public. This is what the new way of technical work looks like — AI-augmented, transparent, process-focused.

Layered note cards and a connected graph illustrate skill-first knowledge systems.

Knowledge Systems Mar 16, 2026 4 min read

You Don't Need Obsidian Anymore (But You Need What It Taught You)

Obsidian taught a generation how to think in links. But the skill — linked thinking, atomic notes, emergent structure — has outlived the tool. The LLM is your …

A clean KPI cover showing proxy and reality lines diverging while DMAIC questions frame the measurement gap.

Process Improvement Mar 15, 2026 5 min read

All Your KPIs Are Wrong (And That's the Point)

Your KPIs do not measure exactly what you think they measure. Here is why interrogating your metrics—not just refining them—is the real work of DMAIC.

$A clean systems diagram highlighting constraint signals, charts, and decision cards for AI hiring.$

Hiring Systems Mar 14, 2026 5 min read

How to Spot Real AI Systems Thinking in an Interview

Framework fluency is easy to fake. The stronger hiring signal for AI systems roles is whether a candidate naturally reasons about constraints, failure modes, and causal chains.

Sunset over Webster Lake with still water reflecting a dark cloud shelf and bright horizon.

Field Notes Feb 23, 2026 3 min read

Welcome to the Supervised Blog System

How we publish practical AI systems writing with approval gates and factual rigor.

ψ CodingEnvironment Data Analytics · AI-Enabled Systems · Process Engineering