Originally published at hafiz.dev Every SaaS app eventually hits the same question: how do you make one application serve multiple customers with separate data? If you're building with Filament, the answer is closer than you think. Filament ships with a built-in tenancy system that handles tenant switching, automatic resource scoping, registration, and profile management out of the box. But here's
TL;DR: I built the same browser agent twice — once with 500 lines of Python, once with 7 lines of JSON. The second one took 5 minutes. The agent harness layer is becoming the real competitive advantage, not the model. Last month, I built a browser automation agent. Playwright. Custom orchestration. Login handlers. Error retries. Session management. React-aware form filling. Anti-detection scripts.
I got tired of the same three-step content publish loop: write draft → open CMS → paste, format, re-paste, fight the rich-text editor, click publish. Repeat for every environment — staging, then production. For one article, fine. For a team publishing 20+ pieces a month? That workflow is a quiet tax on everyone's time. So I wired up a pipeline that cuts the loop entirely. You commit a .md file to
I finished a seven-part beginner series on building a personal website with Codex. The reader I had in mind is not a programmer. They can use a computer, copy text, open a browser, sign in with Google when needed, and read a page to see whether it says the right thing. That is enough for a first site. The mistake I wanted to avoid was making people learn the whole map before they can move. A norma
Il y a quelques années, au lycée (entre 2022 et 2025), un professeur m'a donné le déclic pour l'informatique. Je passais mes journées sur des forums à décortiquer le fonctionnement des réseaux et de la sécurité. Mais j'ai vite été frappé par une réalité : apprendre la tech aujourd'hui demande souvent de "donner un organe". Il faut une connexion fibre, un abonnement coûteux, et surtout, on laisse s
Originally published on rohitraj.tech UPI fraud hit ₹805 cr in India last year. Cloud APIs leak data. So I built ScamRakshak — fully on-device scam detection. 3-tier inference engine: Gemma 4 LLM — context-aware classification LiteRT — fast pattern model Regex fallback — when battery low Full architecture write-up: https://rohitraj.tech/en/notes/build-on-device-ai-scam-detector-android-gemma Read
After getting frustrated with news apps designed to keep me scrolling forever, I built Trace - an Android app that aggregates news from 100+ sources and uses AI to create 30-second summaries. The app is free, no ads, no subscription. Just news that respects your time. The Problem Every morning I'd open Google News or Twitter and 30 minutes would vanish. Not because I was learning anything useful,
Why Figma MCP Isn’t Enough Why Figma MCP Alone Can’t Guarantee Production-Ready UI — and What Product Teams Must Do Instead Extraordinary results require an extraordinary team. I’m surrounded by people who treat design and development like a mission. They are warriors in the tech trenches, and this win belongs to them. No fluff. No filler. Just the facts on how we shattered our veloci