Because Developers are Awesome

Setting up a project for Agentic AI

April 19, 2026 13 minute read

In my last three articles, I produced a set of recipes and rules for doing AI-first development using Spec-Driven Design. These rules really get you started in how to think about development when the coding (the easy part) is done for you. But how do you get started? When I am setting up a new project, my process has now changed. It starts off the same, but the additional files I need to write before I can start coding have changed. Here is what I do: 1. Scaffolding Inevitably, I hav...

Effective prompts for agentic engineering

April 11, 2026 10 minute read

This is the third and final part of my series on agentic engineering. If you want to read the first two articles, see the following links: Ten rules for spec-driven design Using Jobs-to-be-done to improve your agentic spec In the previous article, I covered the spec, which documents the “what” and the “why” of what you want to build using JTBD as a primary product requirements framework to use. This is recommeneded because the alternatives rely on a conversation between a product ma...

Using Jobs-to-be-done to improve your agentic spec

April 09, 2026 6 minute read

In my last article, I introduced ten rules to help you move from vibe-coder to AI Product Architect by building specs. However, I glossed over a few of the details. The basic problem is that most people write specs for themselves - a human. You aren’t writing for a human, so why do you think it would look the same? In this article, I’m going in depth into using product management techniques to write a spec that agents love. Rule 1 (refined) is “Don’t just tell the agent what to build; t...

Ten rules for spec-driven design

April 07, 2026 6 minute read

Recently, I posted an article on the AI Maturity Model. In that article, I proposed six stages of growth - from tactician or accidental editor to the agentic engineer. The biggest difference between the vibe coder and the agentic engineer is the transition to spec-driven design. But what does that really mean? The “vibe-coder” sends one massive prompt and hopes for a miracle. The agentic engineer understands that an LLM is a reasoning engine, not a magician. Spec-driven design provides ...

Are we in an AI bubble or an AI revolution?

April 05, 2026 3 minute read

I read a lot on the Internet from the prognosticators about whether AI is terrible or awesome - are we in a bubble or a revolution in the way we work? Why not both? Perhaps the best way to explain this is by - as everyone else is - drawing parallels to the other bubble / revolution that we all experienced - the dot-com crash of 2000. The year 2000 represents the moment where the financial excitement (the bubble) outpaced the actual infrastructure (the revolution). The Bubble: Financial ...

The AI Maturity Model

April 03, 2026 7 minute read

I have a confession to make. I don’t write code any more, and I haven’t written any code since August of last year. I wrote a little about my journey to AI nirvana (see my posts on AI Editors, OSS AI Editors, or SpecKit). I’m guessing that everyone goes through the same journey - from distrusting that AI will do a good job to using AI for everything. The good news is that software development has not really changed. The job was never about the code, even though the code was the tangible ...

Using SpecKit with multiple AI agents

December 06, 2025 8 minute read

I wanted to put down my workflow and all the details for using SpecKit using a superior reasoning and planning LLM (like Opus 4.5 via Claude Code, or Gemini 3 via the Gemini CLI) for the initial phases, but switching to an agentic AI IDE like Cursor or GitHub Copilot in VS Code for the final coding phases. I’m using Git and the GitHub CLI for repository actions. This workflow does not include “authenticating”. You still need to configure and log in to each tool separately. They’ll general...

AI Assisted Editors: A Comparison (Part 2)

August 01, 2025 10 minute read

In my last article, I provided a break down of the three commercial AI assisted editors that I use on a regular basis. However, not everyone can afford a monthly subscription ($20/month is the standard), or maybe you prefer different models or a more local development environment. There are still options that you can use to get AI assisted editing yet avoid the subscription charge. What is an AI Assisted Editor? As a developer, you open up an editor and write code. AI Assisted editors ...

AI Assisted Editors: A Comparison (Part 1)

August 01, 2025 13 minute read

You can’t go a lap about development these days without bumping into AI assisted editors. Most of the AI tooling companies I see are heavily investing in Visual Studio Code based editors, but you have some others out there as well. I’m going to break the comparison I’m doing into three articles - one on the paid offerings I’ve personally used, one on the “free” offerings I’ve used, and one on the quirks of the LLMs that are in common use. Hopefully, by the end of this, you’ll understand th...

Customizing Keycloak with Aspire - Part 3

October 04, 2024 6 minute read

This is the last article in a trio of articles about incorporating Keycloak into .NET Aspire applications. Thus far, I’ve covered the basics and some common customizations. This part covers what you need to do if you want to store the identity information within a PostgreSQL database. I’ve chosen PostgreSQL for this because it seems to be the database of choice for Aspire when you want to run things locally and in the cloud and also because the PostgreSQL database driver is included by def...

Adrian Hall