R
0
Full Stack Commerce EngineerGreater Delhi Area, India

RajeshGandhar.

I build commerce systems that actually ship.

Specializing in eCommerce platforms, marketplace systems, and full-stack product engineering — from product catalog to checkout.

scroll
React 19Next.js 16TanStack StartTypeScriptPostgreSQLSupabaseRow-Level SecurityPostGISRazorpayZodTailwind CSSVitestPlaywrightGitHub ActionsFramer MotionThree.jsGSAPVercel
React 19Next.js 16TanStack StartTypeScriptPostgreSQLSupabaseRow-Level SecurityPostGISRazorpayZodTailwind CSSVitestPlaywrightGitHub ActionsFramer MotionThree.jsGSAPVercel

Selected Work

Real-world products, not demo apps.

Every project here is a real system built to solve a real problem.

VedaGlow storefront showing the 28-Day Skin Reset kit with pricing and an add-to-cart action

D2C eCommerce Platform

VedaGlow

Live

VedaGlows is a live D2C Ayurvedic skincare brand and eCommerce platform I built end to end — frontend, server functions, database schema, payment integration, admin tooling and production deployment. It sells the 28-Day Skin Reset Kit directly to consumers in India, and handles the full commercial path from landing page to delivered order.

  • ↗Razorpay payments with server-side signature verification
  • ↗Split advance-payment and cash-on-delivery flow
  • ↗Six-state order lifecycle with refund-aware payment status
  • ↗27 Row-Level Security policies across the schema
React 19ViteTanStack StartTypeScriptTailwind CSS v4Radix UIZustandTanStack QueryZodSupabasePostgreSQLRazorpayVercel
GharBazaar homepage with a property search bar offering buy, rent, commercial and plot modes, plus filters for property type, budget and BHK

Property Marketplace Platform

GharBazaar

In Development

GharBazaar is a property marketplace built around a single idea: students looking for housing care about distance to their campus, not postal boundaries. Properties are linked to nearby universities with distances computed in the database using PostGIS, so a student picks their university and sees genuinely nearby flats, PGs and hostels ranked by real distance. Around that sits a full marketplace — tiered identity verification, listing moderation, lead capture with contact gating, in-app messaging, subscriptions and billing.

  • ↗PostGIS proximity search with trigger-computed campus distances
  • ↗58-table schema hardened by 128 Row-Level Security policies
  • ↗Three-tier KYC verification with private document storage
  • ↗196 unit tests and a passing production build
Next.js 16React 19TypeScriptTailwind CSS v4shadcn/uiBase UISupabasePostgreSQLPostGISGoogle MapsVitestPlaywrightGitHub ActionsVercel

Commerce Engineering

What I understand about commerce systems.

Not just features — systems. Each domain represents real engineering decisions.

Built
Experienced
Learning
Planned

Product Catalog

Built

SKU-keyed products with admin-editable pricing, stock and visibility.

VedaGlow

Cart Engine

Built

Persistent cart state with server-side re-validation before an order exists.

VedaGlow

Checkout Flow

Built

Multi-step checkout with PIN-code address auto-fill and server-authoritative totals.

VedaGlow

Payment Gateway

Built

Razorpay with server-side signature verification, plus split advance and cash on delivery.

VedaGlow

Order Management

Built

Six-state lifecycle with independent, refund-aware payment status.

VedaGlow

Coupon & Pricing Engine

Built

Percentage and flat discounts with expiry, activation and global usage caps.

VedaGlow

Authorization

Built

155 Row-Level Security policies — the database is the final authority on access.

VedaGlowGharBazaar

Admin & Operations

Built

Role-guarded admin surfaces for orders, catalog, moderation and settings.

VedaGlowGharBazaar

Search & Discovery

Built

PostGIS proximity search, filtering and saved searches with alert matching.

GharBazaar

Marketplace Systems

Built

Multi-role access, tiered KYC verification, listing moderation and lead gating.

GharBazaar

Inventory System

Learning

Stock reservation, atomic decrement and overselling prevention under concurrency.

Returns & Refunds

Learning

Automated refund processing, reconciliation and inventory restoration.

Experience

Where I've built.

Founder & Full Stack Commerce Engineer

Current

VedaGlow

Built and run a live D2C skincare eCommerce platform end to end — storefront, database schema, payment integration, order operations and admin tooling.

React 19ViteTanStack StartTypeScriptTailwind CSS v4Radix UIZustandTanStack QueryZodSupabasePostgreSQLRazorpayVercel
March 2026 — Present

Independent Project — Full Stack Product Engineer

Current

GharBazaar

Designing and building a property marketplace with university-proximity search for student housing, tiered identity verification and database-enforced authorization.

Next.js 16React 19TypeScriptTailwind CSS v4shadcn/uiBase UISupabasePostgreSQLPostGISGoogle MapsVitestPlaywrightGitHub ActionsVercel
2025 — Present

Web Development Intern

Next Gen Software Hub Pvt. Ltd.

First professional development experience — contributed to web application features and worked within a team delivery process.

HTMLCSSJavaScriptReact
March 2025 — April 2025

Engineering Roadmap

What I'm building next.

These are planned projects — not yet built. Listed here to communicate direction, not inflate claims.

Multi-Vendor Commerce Platform

Planned

A production-grade multi-vendor marketplace with vendor onboarding, dashboards, product catalog, SKU management, inventory, orders, commissions, and analytics. The goal is to understand and implement the full complexity of a real marketplace architecture.

  • ·Vendor onboarding, KYC, and dashboard
  • ·Product variants and SKU management per vendor
  • ·Order routing and vendor-level fulfillment
  • ·Commission calculation and payout system
  • ·Admin controls: approval, moderation, analytics
  • ·Vendor isolation at the data and API layer
MarketplaceMulti-VendorCommerceSystem Design

AI Commerce Engine

Planned

An AI-powered commerce layer with natural-language product search, personalized recommendations, review summarization, and RAG-based product knowledge. Exploring how LLMs and vector search can improve commerce discovery and conversion.

  • ·Natural language product search (semantic + hybrid)
  • ·AI-powered product recommendations
  • ·Review summarization via LLM
  • ·RAG-based product knowledge base
  • ·AI shopping assistant
  • ·Product comparison engine
AICommerceRAGLLMVector SearchSemantic Search