4 years · 4 companies
Every line I've
shipped.
Full engineering history — system decisions, impact numbers, and the products I built. Not a summary.
CheckMe Commerce Pvt. Ltd.
Senior Frontend Engineer · Founding Engineer · Ahmedabad, GJ
Platform Architecture
Multi-tenant SaaS with feature flags, RBAC across 9 role types, 7 order types (Virtual Demo, In-Store Demo, Physical Installation, Self-Service, Training, Consultation, Rental), all on a 60+ table PostgreSQL schema.
AWS Infrastructure
Full AWS setup in ap-south-1: ALB with SSL termination, WAF rules for DDoS/SQLi/XSS, Multi-AZ RDS PostgreSQL, S3 + CloudFront, SES for transactional emails, Route53 — provisioned without a DevOps team.
Embeddable SDK
CDN-distributed single-JS-file SDK embeddable on any website. Supports 5 embed modes (inline widget, modal, floating button, checkout overlay, standalone page) with HMAC-signed session-token auth.
WebRTC Demo Infrastructure
Integrated ZegoCloud for virtual demos — room creation, participant management, recording, and quality monitoring across 6,000+ live sessions.
Brand Onboarding
Led technical onboarding for SharkNinja, Nester, Tiron, EcoFlow — mapping product catalogues, SKUs, demo scripts and CDE training into the platform. Achieved 25%+ conversion vs industry average of 8–12%.
Key Contributions
- Sole engineer for the first 4 months — architected the entire platform from scratch across web, API, and cloud infrastructure.
- Built 10+ Next.js frontends and 5 FastAPI microservices on PostgreSQL Multi-AZ RDS (60+ tables), covering 7 order types and RBAC across 9 role types.
- Onboarded global brands SharkNinja, Nester, Tiron, EcoFlow — achieved 25%+ demo-to-purchase conversion rate.
- Engineered ZegoCloud WebRTC video infra, CDE training LMS with YouTube sync, and multi-channel notifications.
- Designed full AWS infra: ALB, WAF with DDoS/XSS/SQLi rules, Multi-AZ RDS, S3, SES, Route53.
- Built CDN-distributed embeddable SDK (single JS file) with 5 embed modes and session-token auth.
Stack
MathonGo
SDE-1 · Frontend Engineer · Bengaluru, KA
CRA → Vite + TurboRepo
4-minute cold builds on aging CRA (Node 16). Migrated to Vite + TurboRepo — cold builds under 17s (93% faster), page loads -50%, bundle -21%. Zero downtime, no breaking changes.
JEE Main College Predictor
Sole front-end engineer for the tool. Aggressive SEO — structured data, Core Web Vitals tuning, dynamic OG images. 400K users in 48 hours, ranked #2 on Google — fastest traffic event in MathonGo history.
Quizrr Platform Overhaul
Rebuilt test-taking platform from ground up. Cut 11,000 lines of legacy code. Rebuilt freemium gating, Test Analysis dashboard, Onboarding, and Question Review. Lighthouse went from ~55 to 95+.
A/B Testing & Growth
4 WebEngage A/B experiments — different CTA copy, onboarding steps, test completion nudges. Results: 28% rise in test completions, 35% rise in Test Analysis engagement.
Key Contributions
- Proposed and led migration from CRA (Node 16) to Vite + TurboRepo — 93% faster builds, 50% faster page loads, 21% smaller bundle.
- Frontend overhaul directly contributed to platform growth from 1M → 2.5M students.
- Built JEE Main College Predictor — 400K+ users in 48 hours, ranked 2nd on Google Search.
- Integrated WebEngage A/B testing — 28% rise in test completions, 35% rise in Test Analysis engagement.
- Achieved 95+ Lighthouse scores across Performance, SEO and Best Practices.
Stack
Codebuddy Pvt. Ltd.
React Developer · Kolkata, WB
Freelance Marketplace
User profiles, gig listings, search/filter, proposal system, payment integration. Delivered in 6 weeks.
Car Rental Application
Booking system with availability calendar, multi-step checkout, admin panel. TTFB reduced from 1.3s to 750ms via Next.js SSR + SWR caching.
Engineering Practices
6-hour PR turnaround, 5+ production deploys/week with zero rollbacks. SWR caching strategy eliminated stale data across both platforms.
Key Contributions
- Delivered two production client platforms in 3 months — Freelance Marketplace and Car Rental App.
- Next.js SSR + SWR caching: 45% faster load times, zero stale data. TTFB reduced from 1.3s to 750ms.
- Maintained 6-hour PR turnaround with 5+ production deploys per week.
Stack
FWC Inc.
UI Developer · Bengaluru, KA
HR SaaS Dashboard
3 role-specific dashboards with 20+ screens and 40+ API integrations — leave management, payroll, recruitment pipeline, attendance, performance reviews.
RBAC System
Client-side RBAC with permission matrices per role. Reduced role-based bugs by 45%, made adding new roles safe.
Real-Time Visualisations
8+ Chart.js charts with Firebase real-time listeners for headcount, attrition, recruitment funnel, and performance distribution — all without page refresh.
Performance
Code splitting, lazy loading, image optimisation brought page load from 3.4s to 1.8s (47% improvement).
Key Contributions
- Built HR SaaS frontend — 3 dashboards (Admin, Manager, Recruiter), 20+ screens, 40+ REST API integrations.
- RBAC implementation reduced role permission bugs by 45%. Page load optimised from 3.4s → 1.8s.
- 8+ real-time Chart.js + Firebase data visualisations for company performance tracking.