Cover image for I designed and built the

I designed and built the Pristine Admin console, the back-office application that controls all master data behind a manufacturing plant's production system. The challenge was that master data in a factory setting is unforgiving. A mistyped grade name or a deleted material silently breaks months of historical production records, so the panel had to make the right action easy and the destructive action impossible. I built it as a deliberately constrained interface: records can be deactivated but never deleted, every write is validated on both sides, and access is restricted to admin roles at the authentication layer rather than just hidden in the UI. The console manages seven resources across the plant: ball mills, ball mill grades, raw material grades, materials, suppliers, users, and raw material recipes. The recipe builder was the most interesting piece. Each raw material grade needs a formula defining how many kilograms of each material go into one batch, so I built a nested form that pulls live material options, enforces a single recipe per grade by filtering out grades already configured, and validates quantities inline. Because the backend keys recipes by grade rather than by their own ID, I mapped the resource identity in the data layer so the standard list, create, and edit flows work transparently over a non-standard endpoint shape. Most of the engineering went into the data provider. The existing API returns enveloped responses and full collections rather than paginated ones, so instead of reshaping a working production API I wrote a provider that adapts to it: a client-side query engine handling full-text search, boolean filtering, multi-field sorting, and pagination, plus a per-resource field allowlist that strips any key the backend validators would reject before a request goes out. Delete operations throw by design, pointing the user to deactivation instead. Authentication runs through the same JWT service as the operator app, but the console rejects any account without an admin role at login and re-verifies that on every route change, clearing the session automatically on a 401 or 403. I finished it with a custom MUI theme and layout: an indigo and teal palette, a gradient-washed background, softened surfaces and radii, and a collapsible sidebar reworked to show clean centered icons rather than clipped labels when narrowed. Tech Stack: React 19, TypeScript, React-Admin 5, MUI 7, Tailwind CSS v4, Vite, Vercel.

0

25

I worked as a Software Engineer on DripShop, a large-scale live-streaming e-commerce platform that enables creators and businesses to sell products through interactive live video experiences. My work focused on building production-ready features that supported high-traffic user journeys while maintaining performance, reliability, and a seamless shopping experience. The platform combines live video, real-time user interactions, payments, inventory management, and order processing into a single experience. My responsibility was to translate complex product requirements into scalable, maintainable frontend and full-stack solutions, collaborating closely with designers, backend engineers, product managers, and QA throughout the development lifecycle. I contributed to the development of customer-facing experiences across the platform, building reusable React and Next.js components, integrating backend APIs, implementing responsive interfaces, and improving application performance. Alongside feature development, I resolved production issues, optimized existing workflows, and helped ensure new releases met high standards for usability and reliability. Working on a product used by thousands of shoppers required careful attention to performance, state management, and code quality. I participated in code reviews, collaborated on architectural decisions, and followed modern engineering practices to keep the application scalable and maintainable as new features were introduced. This project strengthened my experience building production-scale web applications, working within a large collaborative engineering team, and delivering features that directly impacted real users and business outcomes. Tech Stack: React, Next.js, TypeScript, Node.js, GraphQL, REST APIs, Tailwind CSS, Git, GitHub Actions, Vercel, Agile Development.

0

40

Cover image for I designed and built Pristine's

I designed and built Pristine's entire digital operations platform, from database architecture and role-based workflows to a production-deployed web suite and an AI analytics layer. The challenge was replacing a ceramic processing plant's paper-and-spreadsheet production logs with a system that floor operators would actually use, while giving management the traceability and insight the paper trail never could. Every decision focused on data integrity, least-privilege security, and keeping the interface simple enough for shift operators to work quickly. I built four services around a single PostgreSQL schema. The operator app handles the full production lifecycle: raw material intake, drying plant shifts, multi-section ball mill batch entry, quality control, and finished-goods inventory, with Excel export throughout and navigation that adapts to each of eight distinct plant roles. A modular Express and Prisma API enforces that access control at the route level. A separate React-Admin console keeps master data (mills, grades, recipes, suppliers, users) out of the operator's way. The most important piece was traceability. I designed a quality control gate that sits between production output and inventory. QC records are created automatically when a batch completes, reviewed by an inspector, and only generate an inventory entry if quality passes. Every finished bag carries an auto-derived lot number encoding its production date, flow rate, monthly sequence, and bag count, so any outbound shipment can be traced back through inventory and QC to the exact mill batch that produced it. On top of that I built an AI analytics agent that lets managers query the plant in plain English, architected around a zero-data-to-LLM guarantee. The model only ever sees the database schema and the question, never a row of production data. Five independent layers enforce it: two least-privilege Postgres roles separating read-only queries from insert-only auditing, a sqlglot AST guard that rejects any non-SELECT statement, a per-role table allowlist, the prompt design itself, and a full audit log of every interaction. Answers stream to the browser over Server-Sent Events through a LangGraph pipeline that routes, guards, plans the query, executes, then explains the result in business terms. Tech Stack: React 19, TypeScript, Ant Design, TanStack Query, Tailwind CSS v4, Node.js, Express, Prisma, PostgreSQL, React-Admin, MUI, Python 3.12, FastAPI, LangGraph, OpenAI, AWS EC2, Nginx, Supabase, Vercel.

1

65

Cover image for I designed and built ApexSutra's

I designed and built ApexSutra's entire digital presence—from branding and design system to a production-ready website. The challenge was to create a website that demonstrates engineering quality rather than relying on generic marketing. Every decision focused on performance, scalability, maintainability, and user experience. I developed a reusable design system powered by Tailwind CSS v4, with design tokens for colors, typography, spacing, and motion. The website is fully data-driven, allowing new services and industries to be added through typed content files that automatically generate pages, SEO metadata, sitemap entries, and search integration. SEO is built into the architecture with structured data, Open Graph metadata, and an auto-generated sitemap, ensuring content and search optimization always stay in sync. Performance was equally important, using CSS-first animations, server-rendered content, and accessibility features like reduced-motion support. Tech Stack: Next.js 16, React 19, TypeScript, Tailwind CSS v4, Vercel.

0

64

Bhargav Dabhi

Building Scalable Software That Solves Real Business Problem