I am a forward-thinking Software Engineer and AI/ML Specialist with a product engineering mindset. Currently pursuing a B.E. in Computer Science & Engineering at Sona College of Technology, I specialize in building highly scalable, reliable full-stack applications and integrating machine learning workflows to solve real-world problems.
- Software Engineering: Architectural design patterns, Node clustering, concurrency handling, and relational/non-relational database schemas.
- AI/ML & Deep Learning: Natural Language Processing, Predictive Analytics, Geospatial Clustering, and Prompt Engineering.
- Full Stack Development: End-to-end SDLC ownership, building responsive user interfaces, role-based security, and performance-optimized database structures.
- Product Mindset: Translating complex system requirements into seamless user-facing features, with an emphasis on performance and clean, maintainable systems.
Open To:
- Software Engineering & Full-Stack Development roles
- AI / ML Engineering opportunities
- Open Source and enterprise-scale contributions
| Domain | Proficiency | Details |
|---|---|---|
| Machine Learning & Deep Learning | Advanced | Classification, regression, ensemble models (LightGBM, XGBoost, CatBoost), cross-validation, and Scikit-Learn pipelines. |
| Natural Language Processing | Intermediate | Integrating LLMs (Google Gemini), parsing unstructured textual queries into structured schemas, and Prompt Engineering. |
| Geospatial Analytics & Clustering | Intermediate | Density-based spatial clustering (DBSCAN), GIS data preprocessing, Folium visualization, and urban planning analytics. |
| Generative AI | Advanced | AI-assisted medical triage bots, intelligent booking assistants, automated workflows, and Generative AI for Data Science. |
🩺 Sanguis AI — Blood Donation Intelligence Platform
A machine-learning-driven healthcare logistics solution built to address blood supply chain inefficiencies. It automates donor eligibility evaluation and translates unstructured emergency requests into structured, actionable orders using state-of-the-art NLP.
| Metric | Details |
|---|---|
| Stack | Python · FastAPI · Scikit-Learn · Google Gemini NLP · React · MongoDB · JWT Auth |
| Scale | Handles high-frequency medical queries; structured data schema with indexing for fast lookups. |
| Performance | Instant eligibility predictions; high-speed LLM processing replacing brittle rule-based routing. |
| Security | JSON Web Token (JWT) authorization; secure patient EMR processing and routing logic. |
| Impact | Optimized matching of donors to active urgent surgeries, minimizing response time. |
| Repository | Access Repository |
Sanguis AI integrates machine learning and NLP into a full-stack dashboard. A Scikit-Learn model trained on the UCI Transfusion dataset classifies donor eligibility in real-time. Unstructured text (e.g., chat/SMS queries like "Need 2 units O- at City Hospital") is dynamically parsed into structured objects via Google Gemini NLP, improving processing speed and accuracy.
🚦 Gridlock — AI-Driven Traffic & Parking Intelligence
An AI-powered geospatial mapping and analytics application designed to optimize traffic demand and identify vehicle congestion hotspots across urban zones.
| Metric | Details |
|---|---|
| Stack | Python · Scikit-Learn (DBSCAN) · LightGBM · XGBoost · CatBoost · pandas · Folium |
| Scale | Scaled to handle dense municipal geospatial datasets; cross-validated triple ensemble architecture. |
| Performance | Achieved 90% accuracy in congestion hotspot identification and traffic forecasting. |
| Security | Secure data preprocessing pipelines; anonymized and sanitized coordinate mapping. |
| Impact | Placed in the Top Team out of 30,000+ entries in Flipkart Hackathon. |
| Repository | Access Repository |
Gridlock uses unsupervised machine learning (DBSCAN clustering) to analyze coordinate density and automatically identify zones experiencing severe traffic congestion. Time-series data is processed through a robust triple ensemble pipeline of LightGBM, XGBoost, and CatBoost with 5-fold cross-validation to predict rides/delivery demand. Insights are compiled into an interactive map dashboard built with Folium.
🏥 Telemedicine Platform — AI Healthcare System
A comprehensive clinical management dashboard facilitating patient registration, appointment scheduling, electronic medical records (EMR), and automated symptom triaging.
| Metric | Details |
|---|---|
| Stack | React.js · Node.js · Supabase (PostgreSQL) · AI Chatbot · Health Metrics APIs |
| Scale | Unified multiple health monitoring APIs into a singular, structured database layer. |
| Performance | Responsive UI under load, sub-second latency for API requests and telemetry updates. |
| Security | Role-Based Access Control (RBAC); secure storage of patient credentials and history. |
| Impact | Recognized as a Nasscom Finalist (Top Team / 5,743 entries). |
| Repository | Access Repository |
Designed to streamline digital patient care, this platform incorporates an AI-driven chatbot for initial symptom triage. The application features full EMR functionality, secure authorization, and integrates external medical hardware metrics via APIs. The project stood out for its robust UI/UX design and mobile-responsive layout under heavy live testing during the Nasscom competition.
🏨 Holiday Hideout — Full-Stack Booking Platform
A luxury vacation booking web application supporting role-based logins, booking states, and custom multimedia management.
| Metric | Details |
|---|---|
| Stack | React.js · Node.js · MongoDB · REST APIs · Cloudinary CDN · AI Chatbot · Role-Based Auth |
| Scale | Enterprise booking structure supporting millions of listings; high-throughput image hosting. |
| Performance | 65% frontend performance optimization; 1.5s average page load via database indexing. |
| Security | Role-based authentication (guest/host) and secure session middleware. |
| Impact | Handled full SDLC from wireframing to production deployment with a custom customer service bot. |
| Repository | Access Repository |
Holiday Hideout implements the MVC architectural pattern to deliver a robust reservation dashboard. By utilizing Cloudinary CDN for smart media loading and establishing precise MongoDB database indexing on search criteria, average page loading speeds were significantly reduced to 1.5 seconds.
🛒 Nabula — Ultra-Premium E-Commerce Platform
An ultra-premium electronic and laptop e-commerce site featuring a dark-mode glassmorphism design system.
| Metric | Details |
|---|---|
| Stack | React.js · Java · PostgreSQL · Maven · CI/CD · JWT Auth (Access/Refresh Tokens) |
| Scale | Multi-entity relational database design; admin panel with inventory tracking. |
| Performance | Optimized query pipelines for transaction processes and heavy item catalogs. |
| Security | Dual JWT structure featuring short-lived access tokens and secure refresh tokens. |
| Impact | Shipped a highly polished commercial storefront with secure cart checkout pipelines. |
| Repository | Access Repository |
Nabula showcases a high-fidelity design system that includes responsive UI card components, fluid hover transitions, and a secure checkout flow. The Java Spring/Maven backend interacts with a PostgreSQL database, providing scalable product sorting, seller inventory tools, and strict administrative privileges.
June 2026 – Present (Remote)
- Built and deployed an enterprise-level e-commerce dashboard using React, Node.js, Express, and MongoDB.
- Implemented a high-performance backend using Node clustering, enhancing multi-core execution efficiency.
- Configured Mongoose data modeling and race-condition-safe product seeding to prevent double-transactions under high loads.
- Designed responsive client interfaces featuring complex seller panels, checkout pipelines, and JWT session handling.
React.js Node.js Express.js MongoDB JWT Node Clustering Mongoose
May 2026 – Present (Remote)
- Participating in a structured 6-level enterprise program, working closely on production deliverables.
- Architecting and securing task management REST APIs with Node.js and Express.
- Enforcing standard testing procedures and optimizing endpoint response times.
Node.js Express.js REST APIs API Documentation Postman
June 2026 – June 2026 (Remote)
- Participated in an intensive MERN development incubator.
- Engineered and shipped React-based frontend modules and Node/Express backends under tight sprint timelines.
React.js Node.js Express.js MongoDB MERN Stack
2026 (Remote)
- Selected nationwide for a selective virtual cohort focused on practical applications of artificial intelligence.
- Trained, evaluated, and deployed machine learning classifiers on complex industry datasets.
- Collaborated on research documentation assessing model accuracy, precision, and recall metrics.
Python Scikit-Learn Machine Learning Predictive Analytics Data Science
| Recognition | Details |
|---|---|
| Nasscom Hackathon Finalist | Recognized in the Top Team out of 5,743 nationwide entries for the AI-assisted Telemedicine platform. |
| Flipkart Hackathon Top Team | Ranked among the top teams out of 30,000+ entries for Gridlock traffic demand forecasting. |
| Srinivasan Ramanujan Mathematics Competition | Achieved excellent ranking in the National Mathematics Competition (2024). |
| Samitha Hackathon Recognition | Acknowledged for outstanding project architecture and AI-chatbot integration. |
| Deloitte Tech Job Simulation | Completed advanced software engineering and technical consulting simulations. |
Data Structures & Algorithms in Java — Apna College (Industry Curriculum aligned with Oracle Core Java)
Generative AI for Data Science — IBM (Coursera)
Generative AI Unleashed
Computer Vision 101
Artificial Intelligence Primer
Introduction to Robotic Process Automation
learning: "System Design, Microservices, and Advanced Deep Learning Architectures"
building: "Sanguis AI (Blood Donation Intelligence Platform) & Nabula E-Commerce"
exploring: "RAG Systems, Vector Databases (Pinecone/Milvus), and LLM Agents"
open_to: "Full Stack Developer Roles & AI/ML Engineer Internships/Contracts"