ForestSharing insights on learning, data-driven innovation, AI thinking and my life
AI coding assistants forget context between sessions, reintroduce fixed bugs, and make baffling 'improvements.' Tests are the guardrail that survives the amnesia. Here's what I learned building a real project with Claude Code — and what Kent Beck, Martin Fowler, and the data say about testing in the age of AI.
The Agent Skills spec is open. The example skills are open. But what about the runtime that loads, selects, and orchestrates them? Every major tool — Claude Code, Codex, Copilot, OpenClaw — built their own. Here's what's open, what's not, and what you actually need to build.
Traditional workflow builders like n8n and Zapier connect apps with rigid If-This-Then-That logic. But a new paradigm is emerging: composable AI skills orchestrated by intelligent agents. Here's what's actually working, what's not, and why the shift matters.
My company's internal AI agent builder training was fantastic. But after the demos ended, I started thinking about who these tools actually serve, what they're genuinely good at, and where the gaps still hide. A practitioner's reality check on enterprise AI democratization.
A practical look at how AI is transforming government public services - from tax compliance and fraud detection to taxpayer experience. Real use cases, real ROI numbers, and what IT teams should know about this fast-moving space.
How I moved from spending a euro per blog post on Gemini 3 Pro to a hybrid Gemini 2.5 Flash + Replicate FLUX setup that costs almost nothing — without sacrificing quality where it matters.
I checked Google Search Console and the numbers were embarrassing. Broken hreflang, wasted crawl budget, a sitemap Google refused to read — here are the 12 SEO issues I found in my Next.js blog and how I fixed each one, explained for humans.
Leave a message, say hello!
