A deep neural network built with Keras to estimate the final score of an ongoing IPL innings from live match context.

• PyPython
• KKeras
• DLDeep Learning

• Models current runs, wickets, overs, recent performance, and team information.
• Turns changing match conditions into a practical final-score estimate.

A Next.js application that uses GPT-4 Turbo to generate ATS-oriented résumés and integrates Razorpay for paid access.

• NNext.js
• GPTGPT-4
• RRazorpay

• Combines guided résumé generation with a payment workflow.
• Focuses generated output on applicant-tracking-system readability.

A Flutter productivity application for managing goals, prioritizing work with the Eisenhower Matrix, and reviewing daily productivity.

• FlFlutter
• Dart
• AppMobile Development

• Connects long-term goals with daily task planning.
• Uses the Eisenhower Matrix to make prioritization explicit.

An internal diagnostics web app, built as a Next.js monolith, that reads straight from the database, verifies the main product is producing correct data for every client, and puts the health of all of them on one screen built for troubleshooting.

• NNext.js
• ⚛React
• Docker
• AWS

• Built an internal diagnostics dashboard as a Next.js monolith that verifies per-client data integrity straight from the database.
• Designed the interface for engineers: failures visible at a glance, deep-linkable investigations, and a user's-eye view paired with its API response.
• Deployed on AWS so it reuses the main product's existing single sign-on, with access restricted to authorized members of the engineering team.

A MERN platform for electronic prescriptions that creates new ones, converts existing records, and stores them securely, with multilingual voice input and deployment on AWS ECS.

• ⚛React
• NodeNode.js
• ExExpress
• MongoMongoDB
• Docker
• AWS
• Git

• Integrated voice input and voice commands using the ElevenLabs and Bhashini APIs.
• Set up the project's AWS infrastructure end to end, containerising it with Docker and running it on ECS with CloudWatch alerts.
• Built a GitHub Actions CI/CD pipeline for automated deployment to AWS ECS.

A dental imaging system pairing a C++ desktop application that drives an intraoral camera and detects caries with a Next.js web application I built from scratch for recorded video, live streaming, appointments, and diagnoses.

• C++
• AWS
• NNext.js
• ⚛React

• Worked inside a production C++ desktop application to add AWS S3 video storage and AWS IVS live streaming for an intraoral camera feed.
• Built the companion web application from scratch as a Next.js monolith with AWS Cognito authentication, appointment booking, and diagnosis entry, and deployed it on AWS Amplify.
• Connected live capture to patient records so streamed examinations are stored against the right patient automatically.

A three-part automation system that lets a user drive and automate their desktop from their phone: a mobile-first Next.js web app, a MERN cloud relay, and a Python desktop agent, with live step-by-step visibility and mid-run steering.

• NNext.js
• ⚛React
• NodeNode.js
• ExExpress
• MongoMongoDB
• PyPython
• Docker

• Built a three-part system — mobile-first Next.js web app, MERN cloud relay, and Python desktop agent — for driving desktop automation from a phone.
• Turned prompt or voice input into tasks that are relayed to the desktop and executed there.
• Streamed every step of a run back to the phone and allowed the user to steer the task midway.

A Next.js platform built around care circles, letting family caregivers book appointments and diagnostics for a patient, receive updates from doctors and labs, hold audio and video consultations, and share posts with per-post visibility — deployed on AWS ECS with automated CI/CD.

• NNext.js
• ⚛React
• RTCWebRTC
• RRazorpay
• Docker
• AWS
• Git

• Built a care-circle model letting caregivers act for their patients, book appointments and diagnostics, and receive instant updates from doctors and labs.
• Implemented a chat interface with WebRTC audio and video calling for doctor consultations.
• Added per-post visibility control, Razorpay payments, and an admin dashboard covering the platform.
• Containerised the platform with Docker, deployed it on AWS ECS, and implemented CI/CD with GitHub Actions.

This site. A content-driven Next.js portfolio with its own scroll-choreographed animations, an interactive skill graph, and a CI pipeline that gates every change on real tests rather than on trust.

• NNext.js
• ⚛React
• TSTypeScript
• TWTailwind CSS
• Git

• Built a content-driven site where the case studies, skill graph, and downloadable résumé all read from the same typed JSON.
• Diagnosed an iPad frame-rate regression by measuring rather than guessing, isolating backdrop-filter as the entire cost and cutting frame time from 83ms to 16.7ms.
• Added a real-browser test suite that found a genuine layering bug — a dialog's Close button losing its click to a header control — on its first run.
• Maintain a public status document recording what is still unfinished, including limitations in my own code.