https://juanses.dev/en/blog Notes on architecture, engineering decisions, and things I learn building. en-US https://juanses.dev/en/blog/consistency-over-intensity https://juanses.dev/en/blog/consistency-over-intensity Sat, 27 Jun 2026 00:00:00 GMT You don't learn to code in one heroic weekend. The real roadmap is almost boring: a little, every day, applied and repeated. A story about the 12-hour tutorial that taught me nothing. CraftLearningConsistency https://juanses.dev/en/blog/stepping-away-from-the-keyboard https://juanses.dev/en/blog/stepping-away-from-the-keyboard Fri, 26 Jun 2026 00:00:00 GMT Some problems don't get solved by typing harder. Real rest — a walk, some air, a coffee away from the screen — is a debugging tool, not wasted time. CraftFocusRest https://juanses.dev/en/blog/before-you-start-a-project https://juanses.dev/en/blog/before-you-start-a-project Wed, 24 Jun 2026 00:00:00 GMT The most important decisions on a project happen before the first commit. A short field guide: plan, pick boring tech, ship small, and test in the real world. ProcessPlanningMVP https://juanses.dev/en/blog/great-macbooks-rough-tools https://juanses.dev/en/blog/great-macbooks-rough-tools Mon, 22 Jun 2026 00:00:00 GMT Apple Silicon laptops are excellent. The problem is rarely the machine — it's tools that don't account for it. A concrete example from Android Studio. ToolsmacOSAndroidTooling https://juanses.dev/en/blog/linux-for-development https://juanses.dev/en/blog/linux-for-development Sat, 20 Jun 2026 00:00:00 GMT Not nostalgia — practicality. Configuration that actually works, native Docker, Unix tooling, and raw speed. A working developer's case for Linux. ToolsLinuxDockerDevOps https://juanses.dev/en/blog/billparty-offline-first https://juanses.dev/en/blog/billparty-offline-first Thu, 18 Jun 2026 00:00:00 GMT BillParty has no backend, no accounts, no cloud — on purpose. How a hard constraint (works on a plane, lives on one phone) became the product's best feature. ProductBillPartyArchitectureOffline https://juanses.dev/en/blog/why-i-built-senddock https://juanses.dev/en/blog/why-i-built-senddock Tue, 16 Jun 2026 00:00:00 GMT Transactional email and waitlist tools quietly turn expensive. SendDock is my answer: an open-source platform you can self-host and actually own. ProductSendDockOpen SourceGo https://juanses.dev/en/blog/code-is-read-more-than-written https://juanses.dev/en/blog/code-is-read-more-than-written Sun, 14 Jun 2026 00:00:00 GMT Programming isn't typing. Most of the job is reading, interpreting, and probing code — including your own from last month — to understand a problem before you touch it. CraftReadingLearning https://juanses.dev/en/blog/clever-code-costs https://juanses.dev/en/blog/clever-code-costs Fri, 12 Jun 2026 00:00:00 GMT The cleverest line you write today is the one you'll struggle to read in six months. Optimize for the reader, not for the satisfaction of being clever. CraftSimplicityMaintainability https://juanses.dev/en/blog/tests-as-living-contracts https://juanses.dev/en/blog/tests-as-living-contracts Wed, 10 Jun 2026 00:00:00 GMT A good test suite isn't a safety net you tolerate — it's the executable specification of what your system promises, including the properties that must always hold. ArchitectureTestingDecisions https://juanses.dev/en/blog/make-decisions-reversible https://juanses.dev/en/blog/make-decisions-reversible Mon, 08 Jun 2026 00:00:00 GMT You can't reliably predict the right call. So optimize for cheap mistakes: prefer decisions you can walk back over ones you have to get right the first time. ArchitectureDecisions https://juanses.dev/en/blog/boundaries-before-features https://juanses.dev/en/blog/boundaries-before-features Fri, 05 Jun 2026 00:00:00 GMT Why I draw the lines between modules before writing a single feature — and how it keeps a codebase from rotting as it grows. ArchitectureBoundariesDecisions https://juanses.dev/en/blog/money-as-integers https://juanses.dev/en/blog/money-as-integers Wed, 03 Jun 2026 00:00:00 GMT Floats and currency don't mix. A short note on a decision I make on every project that touches money — and why it keeps balances exact. ArchitectureDomainDecisions