Rate limiting isn't an npm dependency — it's an abuse policy. Before copying middleware, you need to define what asset you're protecting, what abuse pattern you expect, and what a false positive costs you. A guide with a decision matrix, real gotchas, and observability for Next.js.
Jun 22 2026 · 9′ · Tutorials · TypeScript · nextjs
The README on juanchi.dev says "portfolio landing". The code says something else: an editorial system with repo ingestion, quality gate, automatic rewriting, and crons on Railway. The technical story the README doesn't tell.
Jun 22 2026 · 6′ · Next.js · TypeScript
pnpm workspaces is the best option for TypeScript monorepos in 2026. But the happy path in the docs hides three traps that only show up in CI with real deployments: phantom dependencies, broken hoisting on Railway, and script filtering that doesn't filter what you think it does.
Jun 19 2026 · 9′ · Tutorials · TypeScript · pnpm
Kubernetes technical interviews have a problem nobody names: they ask about objects you'll never touch in production, while ignoring the mistakes that actually break real systems. Here's the map I was missing.
Jun 16 2026 · 9′ · Tutorials · docker · devops
A healthcheck that only says "the process is responding" can hide serious business-level failures. Let's break down what the HEALTHCHECK instruction actually promises, where the standard recipe falls apart, and how to use it as the limited operational signal it really is — not as a guarantee of heal
May 28 2026 · 8′ · Tutorials · node.js · docker
Same backend, same database, same k6. In the first runs it looked like Embedded GlassFish was on top. When I fixed JDK, warmup, window, heap, and DB attribution, the story changed: Spring Boot ended up with the best local profile for this workload. Payara Micro was the cleanest Jakarta EE by check failures. GlassFish surprised by being viable.
May 28 2026 · 10′ · Experiments · Performance · backend
Before you install any rate limiting middleware in Next.js, you need to define what asset you're protecting, what abuse you're expecting, and what a false positive actually costs you. The library is the last decision. The policy is the first.
May 21 2026 · 8′ · Tutorials · Next.js · TypeScript
A low p95 with a 97% error rate isn't a fast pool — it's a pool that fails fast. I built a reproducible experiment with Spring Boot 3, PostgreSQL, and k6 to understand which signals actually matter — and which ones deceive you.
May 15 2026 · 13′ · Experiments · Performance · backend
The install-time benchmark I published earlier didn't capture the real cost of pnpm workspaces in CI: silent cache invalidation, dependency hoisting that breaks in App Router, and a specific edge case that can take down your Railway pipeline. Here's what I failed to measure.
May 11 2026 · 10′ · Experiments · TypeScript · pnpm
Architecture, real systems, product decisions and technical lessons. Everything reviewed with my own judgment before it ships.
230 articles published