AddressFix: Clean Messy Indian Addresses with Mappls
Indian ecommerce does not fail only because of ads, pricing, product quality, or checkout design. Sometimes it fails because of one simple line: near hanuman mandir, gali no 3, opposite…
Delhivery Maps + Naksha LLM: A Deep Dive Into India’s First Delivery-Trained Mapping Stack
Every developer who has built for the Indian market knows this problem: our addresses break geocoders. “Near Lajpat Nagar Metro, behind Central Market, gali no 4, Delhi” is crystal clear…
Why I Picked Sarvam Vision Over OpenAI for DocuBharat
In February, I attended the India AI Impact Summit in Delhi. Three days of policy talks, vendor pitches, and a lot of conversation about whether Indian AI companies could compete…
3 Database Mistakes I Made Building a Multi-Tenant SaaS in MongoDB
I spent the last ten months as the lead engineer on a multi-tenant booking SaaS. Node.js, Mongoose, MongoDB Atlas, deployed on Cloud Run. The product is live with real tenants…
Laravel 13: When to Upgrade and What’s Actually Worth It
A working developer’s decision framework — six weeks after release. TL;DR Support Timeline (the only table that matters) Version PHP Released Bug fixes until Security until 11 8.2–8.4 Mar 12,…
Idempotent AI Jobs in Laravel: How to Stop Re-Charging the OpenAI API on Every Retry
A Laravel queue job retried three times last week. OpenAI charged us for all three calls. The user only ever saw one response. Took me a minute to figure out…
Multilingual Embeddings for RAG: Why the Wrong Model Broke Our Hinglish Search
TL;DR — We built a RAG system over 109 podcast episodes (1.6 million words, mostly Hinglish). The first embedding model we tried, a popular English-first one, returned nonsense for queries…
What 3 Production Systems Taught Me About Building for Real Businesses
I’ve been writing code for over 8 years. Laravel, PHP, Flask, Node.js, Python, MySQL — the full stack. I’ve built healthcare recruitment platforms, warehouse management systems, e-commerce portals, and job…
RAG Chunking Strategy for Multilingual Content: How We Made 1.6M Words Searchable
Introduction Most RAG tutorials show you how to chunk a clean English PDF. Real production data is never that simple. When I built Raj Brain — a search system over…
Why AI Projects Fail: 3 Architecture Decisions That Fix It
80% of AI projects fail before production. The problem isn’t your AI model — it’s the architecture decisions you made before you ever called the API. The $547 Billion Problem…