Things worth knowing
AI, full-stack, automation. I write about what I build and what I learn.
Browse by tag
Dev tools — my must-haves in 2026 (terminal, editor, tooling)
After a year of mixing in new tools I have a stable list of dev tools I actually use daily. Showing each one, why, and what I dropped along the way.
Read article
Claude Agent SDK — first contact with building your own agent
The Claude Agent SDK lets you write your own agent on the Claude API with the full tool apparatus. Showing first impressions — what surprises, where the value is, when to pick the SDK over Claude Code.
GitHub Actions vs cron — when to pick which in 2026
GitHub Actions are great for CI/CD and public repos, local cron for anything touching your infrastructure. Showing the rule and concrete cases.
Trading dashboard FastAPI + Next.js — architecture of my setup
I run an internal trading dashboard: FastAPI backend + Next.js frontend, deployed as Docker compose, exposed via Cloudflare tunnel with Zero Trust auth. Showing the structure and key decisions.
Ultrareview vs regular review — when multi-agent makes sense
Ultrareview is multi-agent code review in Claude Code. Showing the difference vs regular review, what it costs, and when it's worth it.
AI cold outreach — the pipeline that works for me
Generating personalized offers for agritourism, carpentry, physiotherapy. AI takes a lead list, runs research, builds demo sites, drafts brand briefs and emails. Showing the pipeline structure.
Spec-Driven Development with Claude Code — one year later, retrospective
A year ago I wrote about SDD as a methodology. After 12 months of heavy use I have data: what actually pays off, what I no longer recommend, and when SDD hurts.
Docker Compose in a homelab — 6 traps that cost me time
Docker compose is my main service orchestration in the homelab. Showing 6 concrete traps I hit over the last year and how to avoid them.
Sub-agents in Claude Code — the orchestration pattern that works
Sub-agents are the key pattern when the main session starts to bloat. Showing when to fire them, what context to pass, and how to avoid the 'shouting into the void' effect.
Caddy reverse proxy — patterns I actually use
Caddy is my default front-line proxy in the homelab. Showing 5 config patterns that work for me — from basic reverse proxy to HSTS preload + auth gating.