Topics

Most posts here come in series, each one picking up where the last left off. Read a topic from the top.

PDLC and the three engineering paradigms 8 posts

The thinking behind PDLC — prompt, loop and graph engineering, from the concepts to a real project end to end.

  1. Prompt Engineering, Loop Engineering, Graph Engineering: What Are They? 2026-08-09
  2. Why Does pdlc-skills Fit All Three Engineering Paradigms So Naturally? 2026-08-13
  3. How Do the Three Engineering Paradigms Interlock in pdlc-skills? 2026-08-23
  4. Getting pdlc-skills Running in Your Own Project 2026-08-30
  5. How pdlc-skills Runs Unattended 2026-08-31
  6. How pdlc-skills Keeps Quality Up When AI Writes the Code 2026-09-01
  7. How pdlc-skills Makes Progress, Change Impact, and Quality Trends Visible 2026-09-05
  8. Running pdlc-skills on a Real Project: Three Features, Start to Release 2026-09-18

Go to the project page →

Getting more out of AI coding 6 posts

Making AI coding cheaper, faster and less draining — saving tokens, tiering models, parallel agents, worktrees, and where you fit in as the controller.

  1. Your docs aren't burning your tokens — your tooling is 2026-06-16
  2. AI coding getting pricier? I cut my tokens by 82% (with real data) 2026-06-17
  3. AI getting dumber the longer you chat? It's not the model—time to take control 2026-06-18
  4. Your AI feels slow? Maybe it's not dumb—you're making it work one thing at a time 2026-06-20
  5. Want AI to work in parallel? First give each one its own workspace 2026-06-21
  6. The AI did the work, but I'm the one who's wiped — you're the controller 2026-06-27

AI memory 3 posts

Giving your AI a memory, pruning it as it grows weeds, and turning the check-up into a tool.

  1. How to make your AI actually get you: give it a memory 2026-07-06
  2. AI Memory Grows Weeds: Why Timely Pruning Matters 2026-07-08
  3. I Built a Health-Check for My AI's Memory: It Diagnoses, It Never Operates 2026-07-13

AI security 1 post

Real and false alarms in AI security — a prompt-injection alert that turned out to be wrong, and more.

  1. My AI cried 'prompt injection!' — and I believed it. Then it turned out to be a false alarm 2026-06-30