On AI systems, agent architecture, information theory, and the strange experience of building software that persists across its own forgetting.
A native Go compiler, no-build-step .ts, import defer, and more. The TypeScript features shaping how I ship in 2026, with real numbers from real repos.
ReadA mental model for local persistence in React Native: the three tiers of memory, which data goes where, and how to stop losing user state on restart.
ReadThe old TypeScript-vs-JavaScript debate quietly ended in 2026. Here is the decision that actually holds up now that TS builds are ~10x faster and Node runs .ts files with no build step.
ReadYour AI agent works in the demo and breaks in production. It's almost never the model. It's these five plumbing failures, and each one has a small, boring fix.
ReadYour agent works in the demo and breaks in production because of one thing: the tool calls. Here's the four-checkpoint architecture that makes tool use reliable.
ReadTypeScript deletes your types before the code runs, so casting an API response is a lie your compiler tells you. Here is why it breaks production and the fix that stops it.
ReadNext.js 16 flipped the framework's defaults: Turbopack builds, dynamic-by-default rendering, and a renamed request layer. Here are the 5 changes that actually affect how you ship, plus 2 I'm still waiting on.
ReadReact Native and Flutter are both production-ready in 2026. Here is the honest boundary between them, from someone shipping four React Native apps to Google Play.
ReadAn agent's process state is not the world state. Why retrying a dead agent double-fires side effects, why success reports lie, and the reconcile-then-retry loop that fixes both.
ReadUse a single call when you can name the steps in advance. Use an agent when the model must decide at runtime. Here is the decision line, with real cost numbers.
ReadNo bridge, Hermes V1 by default, prebuilt iOS binaries, native layout animation, and a real network inspector. What actually changed between RN 0.82 and 0.86.
ReadThe open-source Daylio alternative is now one tap to install. Still free, still offline-first, still no account.
ReadThe LLM API you learned in 2024 got cheaper, longer, and far more reliable. Here are the 5 features that changed how I build, and 2 that still have no clean answer.
ReadPython 3.14 shipped in October 2025. Here are the five features that actually reach your day-to-day code, with runnable examples, plus two things I'm still waiting for.
ReadIt's 3am. The job you set up three weeks ago was supposed to run at 9am every day. You check the logs...
ReadUse TypeScript for code you keep and JavaScript for code you toss. Here is where that line sits in 2026, now that runtimes run TypeScript with no build step.
ReadBearable's free tier cuts you off from data older than 30 days. For an app that only gets useful over months, that's the whole problem. SoulSync is free, open source, and keeps your full history on your phone.
ReadI shipped a "Backup your data" button for an offline mood tracker. It exported every entry to a JSON...
ReadI was one artifact away from a Google Play launch. A signed .aab, that was it. Then eas build -p...
ReadI scheduled a heavy maintenance job to run Saturday at 6am, while I was asleep and nothing else was...
ReadMy paywall presented perfectly. Right design, right layout, right call-to-action. The only problem:...
ReadI was doing a UI cleanup pass on our astrology companion app. One card said "Recurring themes are...
ReadI wanted a mood tracker where the stats weren't behind a paywall and the data wasn't somebody else's problem to lose. So I built SoulSync. Here's the honest comparison.
ReadYour moods are about the most private data you can produce. Here's how SoulSync keeps 100% of it on device, with local SQLite, on-device photos, and JSON export you own.
ReadThe part that helps you is seeing your own patterns. So why is that the exact part most mood apps charge a monthly fee to unlock? The objection, and how open-source plus local-first is the way out.
ReadMy on-device QA used to be an AI tapping my app by pixel coordinates. One regression walk logged 83...
ReadYour store credentials are valid. RevenueCat accepted the service account. The SDK initialized...
ReadI created a Google service account, uploaded the JSON key to RevenueCat, and watched my Android app...
ReadYour test suite is green. TypeScript is happy. The web preview looks perfect. Then you install the...
ReadI run four side projects. A journaling app, an Android app blocker, a healthcare AI tool, and a...
ReadI burned four EAS cloud builds and two hours chasing crashes that had nothing to do with my code. All...
ReadI built an AI assessment app for a consulting firm prospect. Deployed it on Supabase free tier. Sent...
ReadMy CI pipeline was green. Unit tests passed. The APK built and signed without errors. I installed it...
ReadPageviews tell you someone showed up. Custom events tell you what happened next. Here's how I wired PostHog into a real payment funnel and what the data revealed.
ReadI declared two AI agents tied based on 10 test games. The Wilson interval showed I had zero statistical power. Here's the math you should do before making any agent evaluation decision.
ReadI ran 10 games between two AI agents. Agent v3 went 5-5 against Agent v1. I reported "v3 ties v1, no...
ReadGmail's users.watch + Cloud Pub/Sub + Cloudflare Tunnel = sub-5-second email delivery into your agent's filesystem. Total monthly cost: $0.
ReadI missed a prize-notification email by 24 hours because my AI agent only checked Gmail when it...
ReadTwo RLS policies on the same table, glued together by OR, returned every public-flagged row to every authenticated user. Post-mortem of how I shipped it, how QA caught it, and the fix.
ReadA frontend QA pass on a brand-new account opened the library sidebar and saw two notes I had never...
ReadMost journaling apps do not fail on day one. They fail the first week you miss. You skip Tuesday....
ReadI shipped a no-op stub of PostHogProvider.tsx on April 20. I told myself I would come back to it that...
ReadWhy I didn't migrate my 24/7 autonomous agent from Claude Code to Nous Hermes, and the 3 patterns I ported in one afternoon instead.
ReadRunning a first-time-user QA agent before user interviews found 31 screenshots, 11 ranked issues, 4 blockers. All 4 fixed same day. Second pass came back INTERVIEW-READY.
ReadWhy honest self-reflection requires privacy first, and what happens when an AI reads your full written history instead of just today's entry.
ReadA real-world pattern for running platform billing and tenant billing side by side with Supabase auth, RLS, Stripe Checkout, and Stripe Connect.
ReadSwapped Whisper for Gemini 2.5 Flash multimodal on my journaling app's voice capture. Same accuracy, roughly 7x cheaper, one less third-party account, free under existing GCP credits. Full edge function + per-user rate limit inside.
ReadFHIR for Software Engineers (Not Just Healthcare Devs) FHIR (Fast Healthcare...
ReadThe holographic principle says a black hole stores all its information on its 2D surface, not in its volume. AI agent memory works the same way -- and the parallels run deeper than you'd expect.
Read