Journal

Selected essays on applied AI, product delivery, and engineering quality

A curated editorial layer for the portfolio: fewer pieces, higher signal, and a tighter focus on systems, tradeoffs, and proof from real work.

4 selected essaysRSSPolish version

Production RAG in Next.js: What Actually Breaks After Launch

Most RAG demos fail for boring reasons: weak retrieval control, no eval loop, no traceability, and bad trust UX. Here is the production shape I use instead.

#RAG#Next.js#Evaluation#Observability#AI UX

AI Engineering · Feb 20, 2026 · 6 min read

EU AI Act 2026: The New AI Advantage Is the Control Stack

Europe is not just regulating AI. It is rewarding teams that can prove traceability, transparency, oversight, and monitoring in the product itself.

#ai-act#ai-trust#ai-engineering#regulation#transparency#procurement

Applied AI · Feb 17, 2026 · 6 min read

Most AI Products Fail at the Interface, Not the Model

Model quality matters. But users leave when the interface hides uncertainty, blocks recovery, and makes expensive behavior feel unsafe. Here is the AI UX shape I trust instead.

#AI UX#Trust#Human-AI Interaction#AI Product Engineering#Applied AI

AI Engineering · Feb 10, 2026 · 7 min read

Frontend Did Not Die. It Became My Edge in Product AI

I did not move into AI because frontend stopped mattering. I moved because AI products reward the habits demos usually ignore: trust UX, evals, latency, and shipping discipline.

#AI Product Engineering#Frontend#AI UX#Evaluation#Applied AI

Need a quality-first AI product partner?

I work with teams that care about production reliability, SEO performance, and user trust.

Start collaboration See product work