How AI was the Best Bender of My Life

I built guardrails for the code, the architecture, and the security. I forgot to build guardrails for myself. After an eight-month megasprint rebuilding 10 years of work with Claude, I ended up on bedrest for a week. Turns out AI-assisted development is the most potent productivity drug ever created, and the dose that makes it medicine and the dose that makes it poison are closer together than you'd think.

The Closing Window: AI's Value-to-Cost Ratio Is Shifting Fast

In September, I burned through an entire GitHub Enterprise token allocation and barely noticed — the overages cost pennies. I rated that moment at 1,000:1. Value to cost. By February, I'm spending $2,000 a week on similar workflows. The ratio is now 600:1 and falling. The window isn't shut. But it's moving. Here's what I think is driving it, and what I'm planning for when it closes.

People > Machines (The OODA Loop Strikes Back)

Eight years after writing "People > Machines," the thesis holds stronger than ever. AI-assisted development has turned the OODA loop into a double-edged sword: with solid devops practices, it creates virtuous cycles that compound quality at unprecedented speed. Without them, it amplifies technical debt into a death spiral. The microphone-next-to-the-speaker problem is real, and the human in the loop is the one who decides what gets amplified.

Why I'm Not Losing Sleep Over AI Stealing My Job or My Soul

After six months of building WitFoo's analytics platform with Claude Code, I'm more convinced than ever that AI coding tools make experienced practitioners more valuable, not less. Claude is an extraordinary Warrior, but the Wizard's strategic vision, the Poet's search for meaning, and the hard-won knowledge of the WitFoo Way remain stubbornly, irreducibly human. The Foo has been upgraded. The Wit remains ours.

Teaching Claude the Old Tricks: Lessons from Migrating Legacy Code to an AI-Assisted Workflow

A companion to "Coding with Claude," this post covers the four-phase approach we used to prepare legacy code for AI-assisted development: documenting what exists, establishing standards, cleaning up the worst of the mess, and letting Claude build the bridge to next-generation code.

GrrCON 2024 - Birthing Perjury-free AI

Abstract

Cybersecurity analysis leading to deterrence of cybercrime requires processing thousands to billions of digital signals per second. Those signals must be accurately comprehended, forensically preserved then used to detect and investigate potential cybercrime. The work products must not only assist the investigators but must be translated into language that non-technical lay audiences including judges, lawyers and jurors can understand.

This presentation explores how generative artificial intelligence (GenAI), natural language processing (NLP), graph-theory and artificial narrow intelligence (ANI) can play a role in delivering these outcomes.

The session includes demonstrations of opensource toolkits, datasets and models designed to assist in this work.

Subscribe to Artificial Intelligence (AI)