Case studies, client work and earlier projects.
// request flow Flutter app → REST API (Node.js · TypeScript) → retrieve (PostgreSQL + pgvector) → LLM answer + source citations
AI-powered knowledge assistant that enables users to search and ask questions about company documents using RAG.
Inventory and sales platform: a Go/Gin REST API with role-based access control, a Flutter admin app and a Python forecasting service, verified by 1,613 automated tests and CI.
// request flow storefront → GraphQL API (Magento 2) → custom modules: catalog · PO/backorder · billing
Backend work on Magento 2 storefronts for enterprise brands: catalog services, multi-store and backorder logic, and GraphQL APIs. GraphQL optimization cut client-side requests by 15%.
ERP features for inventory and sales: product images, configurable low-stock alerts, and sales-order retracking.
Figma-to-Laravel company profile with WhatsApp Direct Chat, and hero slider.
Warehouse & logistics web app—tracks inbound/outbound, inventory, and delivery documents with role-based controls.
Laravel-based Excel import for products & prices with detailed history logs.
Backend REST APIs for catalog, cart/checkout, orders, and admin/reporting with auth, validation, pagination, and caching—speeding up and making day-to-day activities more reliable.
Backend catalog/search, checkout & orders, and admin/reporting APIs with auth, validation, caching, and pagination.
XML export and GraphQL additions for production date, reviews/ratings, form labels, and spec forms.
Shopware storefront improvements, Excel import to DB, custom footer, and N8N integration.
Backend GraphQL APIs for orders, billing, and customers with scoped auth, caching, and pagination—making day-to-day operations faster and safer.
Catalog backend services (PLP/PDP), store-view/multi-store rules, READY vs PO/backorder logic, and ticketing integration—with good validation, caching, and stable error handling.
Custom after-quote logic, and forgot-password redirect fix using Magento 2.
Flutter analytics app with dashboards, leaderboards, and geolocation.
Predict PM2.5/PM10 from weather features using Linear/SVR + tree/boosting models with cross-validation and tuning.
