transforms problems intoproducts

Pursuing B.Tech in Information Technology (4th Year — Graduating 2026) with a strong foundation in core Computer Science subjects, including Operating Systems, DBMS, Computer Networks, Algorithms & Data Structures, OOP, and AI/ML fundamentals. Built multiple real-world projects throughout the course, strengthening both theoretical understanding and practical development skills. Passionate about applying engineering principles to build efficient, scalable, and user-centric digital solutions.

Worked as AI & Full Stack Developer at Everything about AI (Remote). Built AI-powered apps using generative AI and modern full stack (React, Next.js, Node.js). Explored code-driven video editing with Revideo and contributed to intelligent product features.

Served as Full Stack Developer at Madquick, contributing to end-to-end web application development. Designed and implemented responsive user interfaces using React and modern CSS frameworks while building robust backend APIs with Node.js and Express. Worked on database design and optimization using PostgreSQL and MongoDB, ensuring efficient data management and query performance. Participated in agile development cycles, code reviews, and deployed scalable applications to cloud platforms. Enhanced application performance and implemented secure authentication systems.

Developed AI + ML + Drone based Agricultural Automation System, capable of Actual Evapotranspiration (AET) prediction, water irrigation scheduling, and crop disease detection — all in a single platform. Integrated MODIS & Landsat satellite datasets, machine learning regression models, IoT sensors, Raspberry Pi, and Arduino with MQTT + FastAPI backend for real-time data streaming. Built the web dashboard, sensor communication layer, ML pipeline, and drone imaging workflow, delivering 90%+ AET prediction accuracy and earning national recognition under Hardware Problem Statement 1571( Team Swift avigator)

Designed a bio-inspired metaheuristic algorithm (WOA) modeled on wolverine hunting patterns for high-dimensional feature selection in NIDS. Applied on the NSL-KDD dataset, improving intrusion detection performance across multiple ML classifiers by boosting accuracy and significantly reducing false alarm rates. Achieved Bagging classifier accuracy of 99.775% with FPR 0.25%, outperforming GA, PSO, SOM, and Naïve Bayes in benchmarking. Demonstrated strong research aptitude in optimization, ML model evaluation, cybersecurity, and experimentation at scale.

Participated in a real startup-style MVP simulation, working as CEO within a virtual team to build a product from 0 → 1. Gained hands-on exposure to lean startup principles, market research, user discovery, business model refinement, prototyping, and pitching. Learned to align technology with business strategy, validate ideas with real users, and prioritize features around value instead of assumptions — sharpening both product thinking and entrepreneurial mindset.