Experience graph

Worked on a high-traffic, real-time trading platform used by hundreds of thousands of active traders globally. The frontend acted as a long-running system handling live market data, compliance rules, and performance-sensitive workflows.

• Root-caused a 1GB memory leak on the trading page — a useEffect with an incorrect dependency was re-subscribing to every WebSocket tick, cascading re-renders across the order book. Reduced heap to 400MB.
• Restructured Sentry SDK integration — replaced wildcard imports with tree-shakeable named imports, moved initialization to first paint. Cut monitoring bundle from 400KB to 70KB gzipped (82% reduction).
• Led migration of 500K+ existing users to i the India-specific platform by designing geo-based routing, compliance gating, and seamless fallback flows — zero downtime.
• Built position analytics dashboards consumed by 100K+ active traders, optimized for frequent market data updates with minimal re-renders using Apache ECharts.
• Routed 600K+ new sign-ups across platforms based on jurisdiction using location-aware onboarding, reducing manual intervention and support load.

Built an AI-assisted mental health platform from scratch, owning frontend architecture, backend systems, mobile app, infrastructure, and product flows.

• Built a React Native + Expo mobile app wrapping the web frontend in a WebView with native bridges for FCM push notifications and Razorpay payments.
• Designed chat-based mental health assessments that behave like stateful conversations — 24+ message types with real-time Socket.IO events.
• Architected backend with Node.js, PostgreSQL, Redis, and 47 Knex migrations — strong guarantees around correctness, ordering, and data recovery.
• Deployed full observability stack — Prometheus, Grafana, Loki, Tempo — with alerting on unhandled exceptions, high latency, and memory usage.
• Set up Docker-based infrastructure on Coolify with staging/production environments and EAS cloud builds for mobile.

Worked on a consumer-facing salary insights platform where SEO, performance, and rendering correctness directly impacted growth.

• Built salary and compensation insights features used by 20K+ users with interactive ECharts visualizations.
• Improved Core Web Vitals to achieve sub-1s FCP and LCP through bundle splitting, rendering optimization, and SSR tuning.
• Achieved consistent 100 SEO scores across critical landing pages.
• Built an internal design system using Amazon Style Dictionary with custom transforms, adopted across React Native and web teams.

Frontend engineering at India's largest e-commerce platform, operating at massive scale with a strong emphasis on SEO, SSR, and analytics accuracy.

• Built SSR and SEO features serving millions of monthly users on Flipkart's web platform.
• Migrated analytics pipelines to improve data correctness and reduce reporting inconsistencies across teams.
• Developed internal SEO tooling adopted by multiple teams to standardize optimization practices.