AI Learnings: May/2026 version
We are almost past the first half of 2026 and I want to take a moment to share the learning sources that have been most useful to me and may be useful to others.
The basics: AI literacy
With the EU AI Act rollout, the company I currently work for launched GoodLearn access to all it’s employees. Right off the bat I have completed three courses: Introduction to GenAI, Data Literacy with AI and Visualize Your Data. What I like about GoodLearn is how accessible the content is through the mobile app. It feels like doing Duolingo on a very current and useful topic. If this sounds interesting for your org as well consider reaching out to the GoodHabitz team.
Learning with Claude Code
In the last year, I have switched from ChatGPT + GitHub Copilot to fully using Claude (and Claude Code). During that transition I decided to take on the official Anthropic courses. I strongly recommend anyone using Claude to take a few of these (or all) courses as they cover Claude’s most important foundational concepts and features. These are the ones I’ve taken so far:
- Claude 101
- Claude Code 101
- Claude Code in Action
- Introduction to agent skills
- Introduction to subagents
Matt Pocock (aihero.dev)
Matt Pocock was the creator of the grill-me skill. If you haven’t used /grill-me yet, try it now: it turns Claude into a relentless interviewer that stress-tests your plans until every assumption is surfaced. Furthermore, Matt is an educator who has moved fully into AI and offers both free and paid training on conferences, YouTube and at aihero.dev.
- The AI Engineer Roadmap
- “Software Fundamentals Matter More Than Ever” — Matt Pocock
- Full Walkthrough: Workflow for AI Coding — Matt Pocock
Harness engineering: the term of 2026
Martin Fowler continues to excel both at writing splendidly and at inviting writers to his blog who are equally eloquent about tech. I want to highlight three materials worth a read and study:
- Humans and Agents in Software Engineering Loops
- Harness engineering for coding agent users
- Podcast: What is harness engineering?
Geoffrey Huntley
Geoffrey named and popularized Ralph loops. His posts and videos provide inspiration for what AI-assisted coding can do today and what the future may bring.
In mid-2025 he shared a six-month recap which I still consider worth reading for a broad perspective on the impact and transformations brought by GenAI — particularly his observations on how the role of the developer is shifting faster than most teams realize.