Full-Stack Software Engineer building scalable Web, SaaS & AI applications

ReadGrid is a gamified virtual workspace platform where reading books directly fuels the growth of your personal digital room. Users authenticate through a dual-flow system supporting OTP-based email login (via Resend API) and Google SSO (via Supabase), both synchronized with a custom FastAPI backend that issues JWT sessions. After onboarding — which classifies users into reader archetypes through a personalized survey — users access a Figma-like dashboard where they can pan, zoom, and arrange furniture assets that are unlocked by completing AI-generated reading comprehension quizzes powered by Groq's Llama 3.3 model. Each quiz rewards XP and triggers level-ups, making reading a measurable progression system. The frontend is built with React 18 and Framer Motion for smooth micro-animations and deployed globally on Cloudflare Pages, while the Python backend (FastAPI + SQLAlchemy + MongoDB Atlas) runs in a Dockerized container behind an Nginx reverse proxy, with all secrets managed through HashiCorp Vault. Workspace furniture assets are stored in Cloudflare R2 and served as a zero-egress CDN, with metadata indexed in MongoDB Atlas.

An end-to-end logistics platform that seamlessly connects users with verified drivers, featuring an upcoming OpenAI-powered chatbot for real-time customer support.

A highly interactive, gamified 2D personal portfolio designed around a space exploration theme. Built on React and Vite, the application leverages the Kaplay game engine to render a transparent canvas containing a movable astronaut character alongside a custom HTML5 canvas background animating a dynamic starfield. Users navigate the system by clicking or tapping to guide the character through four thematic sections including About, Skills, Contact, and Projects. The application manages its reactive state using Jotai and implements animated popup modals via Framer Motion. Key integrations include a scheduling modal connected to Calendly, a functional contact form utilizing Formspree that automatically overlays when the player collides with the contact zone, and a customizable camera controller for zoom scaling, all tied together with a toggleable background audio system designed to gracefully handle modern browser autoplay restrictions.

Flappy Bird in 2 Languages is an interactive game replica designed to demonstrate identical gameplay physics, asset rendering, and collision detection across desktop and web browser environments. The desktop version is written in Python using PyGame for window management, audio mixing, and graphics rendering, and is packaged into a standalone Windows executable using PyInstaller. The web browser version is implemented in Vanilla JavaScript, HTML5, and CSS3, leveraging the HTML5 Canvas API to implement the identical game loop, scroll mechanics, and mobile-friendly touch inputs, with deployment managed via Netlify. Both versions share the exact same gameplay parameters—such as gravity, jump velocity, and pipe dimensions—as well as the exact same image and audio assets, featuring a dynamically color-shifting user interface for the start and game-over screens.