Building Apps on Oracle with AI on a Mac
Why SQLcl can be a better fit than MCP when AI coding tools need repeatable Oracle database scripts and build workflows.
Category
AI-Assisted Software Engineering
Practical writing on using AI tools to design, build, test, and maintain software.
AI-assisted software engineering is not just about faster coding. The harder question is how we use AI without losing design judgement, testing discipline, or control of the resulting system.
These posts focus on practical software work: agents, specifications, coding workflows, database access, and the difference between useful acceleration and uncontrolled automation.
AI Agents Coding for Hours — What’s Really Going On?
What claims about AI agents coding for hours really mean, and why useful software still depends on architecture, judgement, and tight feedback loops.
How to Build Good Software with AI — Not How You’ve Been Told
A practical argument for using AI in software delivery through exploration, canonical specs, review, and controlled implementation.
What’s Next for Coders?
How AI coding tools may change the role of developers, shifting more value toward design, judgement, specification, and review.
Building Your First Real GPT Is Not a Prompting Exercise
Building a useful GPT is less about clever prompting than treating it as a small software project: source material, structure, tests, iteration and version control.
What AI Changes About Building Blogs
AI has changed the economics of moving a blog: the hard part is no longer the migration work, but knowing what you want to build.
AI Works Best Before You Start Coding
AI is most useful before coding starts: expanding the problem space, surfacing assumptions, and helping shape better decisions before implementation narrows the options.
Don't let ChatGPT write your white paper — and don't let Claude, either.
Two leading language models pulled a white paper in opposite directions. The useful workflow uses both, then puts human judgement back in charge.