Building scalable, fast web applications — six years deep in frontend craft, and always exploring the things I never got to: new frameworks, new problems, new businesses, fun side projects.
I'm Subhra — a Senior Frontend Engineer with 6+ years building scalable web apps in JavaScript, TypeScript, and React, most recently at Billion Hearts. I work on responsive UI, performance optimization, and the unglamorous details of code quality that make products feel calm — and I'm currently open to new roles and available to start immediately.
Along the way I've led projects across Wayground, Hiver, Original4Sure, and Huawei — shipping AI reporting, automation modules, and performance wins, and mentoring developers to keep delivery maintainable. Outside of shipping product, I spend my time exploring the things I never got to: digging into new frameworks, fresh problem statements, and how new businesses work — then turning the best ideas into small, fun side projects.
Built and scaled the distribution channel for Picsee — a free image uploader that shares photographs 10× faster than Google Drive — from scratch, running growth experiments and tracking analytics to lift conversion.
Shipped an interactive Hot Text quiz type, added localization and Sentry error monitoring, built an AI reporting system that turns natural-language queries into real-time automated insights, and lifted third-party LMS grade-sync accuracy from 94% to 99%.
Led custom fields, custom objects, connectors, and automation — a 9.2% ARR uplift. Cut extension failures from 10% to 0.8%, dropped load time 80% (20s → 4s), and piloted AI to shave 20% off timelines.
Built an anti-counterfeit widget that reduced fake goods by 80%, shipped a digital warranty tool with 1M+ registrations, improved legacy-widget Lighthouse performance by 35%, and launched the ChannelBLU MVP for 50,000 portal users in 30 days.
Built an internal AI training platform with 10,000+ monthly logins at 85% backend / 78% frontend coverage, and improved query efficiency of OpenLooKeng — an open-source big-data engine — by 10% via spilling.
Occasional long-form notes about frontend craft, the new frameworks and ideas I'm exploring, and the small disciplines that compound over a career.
I built a small tool that takes any image, shrinks it, and converts it to WebP — usually 25–80% smaller with no visible loss. Here's how it works, plus a from-scratch primer on what an image really is, so the whole pipeline makes sense.
Quick peeks at what I'm reading and keeping close — the things that keep the engineer human.
I’d love to chat about frontend craft, career transitions, or design‑engineering — no strings, no pressure. Say pretty much anything.