CV

Research Assistant

University of Twente

• Energy Management System

Master

Acharya Narendra Dev College, University of Delhi

Computer Science

• Data structure and algorithms
• Linear Algebra
• Computer Networks
• Database Management Systems
• Operating Systems
• Computer Architecture
• Discrete Mathematics
• Abstract Algebra
• Theory of Computation

Master

Central University of Rajasthan

Computer Science

• Machine Learning
• Combinatorial Optimization
• Advanced Algorithms
• Game Theory
• Data Mining
• Compiler Design

Graph Burning | Research Project

Developed and implemented two novel algorithms addressing the graph burning problem, contributing advancements in parameterized complexity.

• Treewidth + Burning Number: Designed an algorithm using treewidth and burning number as parameters, contributing novel insights to the graph burning literature.
• Distance to Threshold Graph: Improved an existing algorithm by reducing its computational complexity from double exponential to single exponential.
• Published the research in the 34th International Workshop on Combinatorial Algorithms (IWOCA 2023) in Taiwan.

Grocery Web App | Full Stack Application

Designed and developed a grocery price comparison platform enabling real-time price retrieval from multiple supermarkets with a scalable backend.

• Developed backend using Python FastAPI to retrieve pricing data across 5 major supermarkets.
• Integrated search functionality using a Sentence Transformer model deployed on AWS Lambda.
• Built scalable backend infrastructure with AWS EC2 and API Gateway for low-latency responses.
• Automated deployment with CI/CD pipelines using GitHub Actions.

Traveling Salesman Problem Solver | Genetic Algorithms (Open Source)

Developed a Python-based solver for the Traveling Salesman Problem using genetic algorithms, enabling efficient route optimization for complex datasets.

• Implemented core genetic algorithm features including selection, crossover, mutation, and fitness evaluation to enhance solution quality.
• Achieved notable community engagement with 20 forks and 10 stars on GitHub, reflecting its adoption in open-source circles.

Anomaly Detection in Hardware Systems

Developed an end-to-end anomaly detection pipeline to monitor hardware vibrations using machine learning and cloud infrastructure.

• Integrated 50 sensor nodes with Raspberry Pi for data collection and wireless transmission.
• Designed and deployed REST APIs for transferring sensor data to AWS Cloud for centralized processing and storage.
• Built machine learning models (Isolation Forest, Random Forest, Autoencoder) achieving up to 93.7% anomaly detection accuracy across diverse test scenarios.
• Developed a real-time dashboard for visualizing sensor data and anomaly alerts, deploying the complete solution on AWS Cloud for scalability and reliability.