Self-taught developer with a Statistics and Data Science background, driven by curiosity and problem-solving. I specialize in JavaScript and React, and I enjoy turning ideas into working products. Exploring AI has expanded how I think, helping me architect and build end-to-end projects. I’m always learning new things, old things, and deeper fundamentals. The more I learn, the more I realize how much there is left to explore.
✉️ ****[email protected]
Front-End Developer JULY 2023 - PRESENT
Though I joined here as a frontend developer, I sometimes have to handle backend tasks too because I work as the solo developer. While working here, I have built around 4+ medium-scale projects, including a job applicant tracking system, automated invoice document management for accounts payable, and hospital staff and patient task and flow management, and patient check-in application. Some projects are still under development, such as a FHIR application integrating eClinicalWorks and an EOD platform.
YTFCS ATS is an applicant tracking system built to streamline the hiring process for Arista Virtual Pros, a side business of ProviderLINK. They hire virtual assistants for US-based healthcare companies. Manually managing the large number of applications for each job had become cumbersome, clearly demonstrating the need for an ATS application.
The initial start was based on their idea, detailed problem statements, and my research on similar applications. Then I finalized the technical requirements, planned the architecture, and designed the initial data structures and collections based on the required features. These features included client entity setup, job creation, multi-channel applications, resume and video-intro processing, secure data handling, hiring team collaboration, kanban board for job pipelines, on-platform video interviews, internal team messaging, maintaining client and candidate history, and many more.
Following the plan, I used Vite with TypeScript + React. I managed all data using Redux. For styling, Tailwind was my best choice along with the Shadcn component library. After completing the UI with mock data and integrating a Node.js and Express.js backend with MongoDB, I found the experience was below expectations because it was slow and had issues with real-time data transactions. Since adding WebSockets was not an option at that time, I shifted the database to Firestore, using Firestore Admin for all backend operations and snapshot listeners on the frontend for true real-time data fetching.
Even though Firestore is serverless and could technically replace the whole backend, we kept the Node.js server because it handled important integrations like OpenAI for resume parsing and validation, Cloudinary for storage, Zoom for video interviews, Adobe for doc processing, IMAP for emailing, and JWT-based custom auth.
The backend was deployed on Heroku and the frontend on Vercel. It now runs smoothly and has dramatically improved the company’s hiring process, with data accuracy reaching around 95%. This saves time, allowing the team to focus more on bringing in clients instead of managing endless candidate lists.
💡 In this project, I have experimented with GitHub CoPilot Agent + Claude Sonnet 4 and 4.5 models thoroughly.
https://github.com/mtauhidul/kiosk
KIOSK is a React-based healthcare check-in app where pre-booked patients submit their health information before appointments. CSS Modules and MUI are used for styling and responsiveness. The multi-screen flow is managed with Redux, and the entire project is a clean translation from Figma to code. After the API integration and testing were completed, it was deployed to Vercel.
https://github.com/mtauhidul/CareSyncFlow
CareSYNC is a medical patient and staff flow management app built with React and the Context API. It includes RBAC-based staff dashboards using Firebase Auth. Pixel-perfect Figma-to-UI translation. Features include managing rooms, doctors, assistants, receptionists, status, and assignments, all powered by real-time Firestore updates. It helps hospital teams track tasks, monitor every room’s status, and respond quickly to emergencies or assistance alerts.
https://github.com/mtauhidul/niblet-ai-v2
A mobile-first, AI-powered web app for tracking daily calorie intake and weight logs to reach personal health goals. It’s essentially a simple personal health and productivity tool built for a US client on Upwork. The stack includes Next.js, Tailwind, Shadcn, the OpenAI Assistant API, and Recharts, designed as the MVP for a future SaaS product.
https://github.com/mtauhidul/synomedix
An AI-powered web app that analyzes critical care patient data to spot early warning signs and help medical teams act faster. Using sample hospital data, it shows how AI can quickly assess risks for ICU, NICU, and CCU patients and suggest helpful, personalized interventions. It’s a simple React-based application built from my own idea and experimentation.