Tushar image
Harmonizing code and creativity, I'm a Computer Science student orchestrating a symphony in the realms of Machine Learning and Data Engineering.
Resume

EDUCATION

University of Massachusetts Amherst | Master of Science in Computer Science | GPA - 3.95 / 4.0

Sep 2023 - May 2025

- Courses Taken: Algorithms for Data Science, Machine Learning, Intelligent Visual Computing, Information Retrieval, Software Engineering, Computer & Network Security, Business Intelligence and Analytics, Data Science Fundamentals, Internet Law and Policy

University of Mumbai | Bachelor of Engineering in Computer Engineering | GPA - 9.15 / 10.0

Aug 2019 - May 2023

- Courses Taken: Machine Learning, Artificial Intelligence, Data Warehousing & Mining, Database Management

CORE PROFICIENCIES AND INTERESTS

  • Machine Learning
  • Natural Language Processing
  • Information Retrieval
  • Quantitative Analytics & Predictive Modelling
  • Data Engineering
  • Data Visualization and Reporting
  • Full Stack Development
  • Project Management
  • Cloud Computing

TECHNICAL SKILLS

Programming Languages & others: Python, Java, C, C++, SQL, JavaScript, R, Git, GitHub
Databases & Visualization Tools: MySQL, PostgreSQL, Snowflake, Redshift, MongoDB, PowerBI, Tableau, Looker
Full Stack Development: Node.js, HTML/CSS, AJAX, Bootstrap, MySQL, JQuery, PHP, Laravel, Django, Flask
Python Libraries: NumPy, Pandas, Scikit-learn, NLTK, Matplotlib, TensorFlow, Keras, PyTorch, Seaborn
Cloud Platforms & DevOps: AWS (SageMaker, EC2, S3, Lambda, Redshift, RDS, EKS), GCP (BigQuery, Vertex AI, Dataflow, Cloud Functions), Docker, Kubernetes

PROFESSIONAL EXPERIENCE

Full Stack Development Intern | Infovue Solutions Inc.

Aug 2021 - Jan 2022

  • Integrated RESTful APIs with external payroll services and benefits providers as part of an HR automation project, significantly improving data synchronization and reducing manual administrative tasks.
  • Led development of a MERN based internal employee portal to create a dynamic dashboard for HR and payroll data visualization.
  • Assisted in the migration of a legacy monolithic system to a containerized microservices architecture using Docker and Kubernetes, ensuring scalability for 2x projected traffic growth.
  • Implemented horizontal scaling via AWS ECS by configuring auto-scaling policies and optimizing load balancing strategies, enabling the platform to efficiently handle up to 1500 concurrent users with minimal latency and high availability.
  • Architected an advanced permission and role-based access control (RBAC) module, ensuring secure handling of sensitive employee data and maintaining regulatory compliance.
  • Lowered cloud infrastructure costs by 20% through resource utilization analysis, right-sizing EC2 instances, and configuring auto-scaling policies, reducing wasteful AWS spending.
Technology: MongoDB Express.js React.js Node.js CRUD Databases SQL PHP Git GitHub Redux HTML/CSS JavaScript Bootstrap FTP Server Wordpress UI/UX JQuery DataTables Ajax AWS

PROJECTS

Optimizing Self-written Machine Learning Algorithms for Real-World Data Applications

May 2024

  • Conducted a comparative analysis of self-written models (Neural Networks, Random Forests, and KNN) on 4 diverse datasets.
  • Designed and implemented multi-layer architectures with custom back propagation, activation functions (ReLU, Sigmoid), and weight initialization, achieving up to 98.14% test accuracy on the Handwritten Digits dataset.
  • Executed comprehensive k-fold cross-validation on all models, ensuring robust generalization, reduced overfitting, and consistently high accuracy and F1 scores.
Technology: Python ML Neural Networks KNN Decision Trees Random Forests

Generative 3D Reconstruction from Single Image Analysis

Apr 2024

  • Crafted a 3D reconstruction pipeline that generates high-fidelity 3D models from 2D images using depth estimation and OpenAI’s CLIP (Contrastive Language-Image Pre-training) for semantic consistency.
  • Fine-tuned the model by freezing encoder layers and optimizing decoder layers, significantly improving visual quality of the 3D reconstructions with efficient resource utilization.
  • Enhanced texture quality using multi-resolution triplane sampling, boosting performance with no extra computational cost.
  • Decreased computation time and enhanced training efficiency by replacing the original L2 loss with a combination of L1 smooth and cosine loss.
Technology: Deep Learning Python Fine-tuning CLIP FID Score Inception Score

Personalizing LLMs (Large Language Models) based on User Profile

Dec 2023

  • Fine-tuned Google's FLAN T5-base model using user profile data and devised a pipeline to generate personalized prompts.
  • Formulated a query generation function using Lexical Augmentation, elevating accuracy by 44.7% with a 120% F1-score increase.
  • Optimized retrieval with a top-k article extraction method using BM25 retriever.
Technology: Python Deep Learning LLM BM25 Flan-T5 NLP NLTK PyTorch Transformers Query Generation Prompt Engineering Fine-tuning

ML-Based Web Platform for Early Detection of Fatal Diseases

Feb 2023

  • Developed and evaluated multiple ML models (SVM, Random Forest, XGBoost, AdaBoost), using hyperparameter tuning and k-fold cross-validation to identify the best-performing model for each disease.
  • Implemented preprocessing techniques such as feature engineering, SMOTE, and standardization, improving robustness and boosting prediction accuracy of the models by up to 4%.
  • Constructed a Stacking Ensemble model (Logistic Regression, KNN, Decision Trees) with a leading 91.3% accuracy in heart disease prediction among seven ML models.
  • Assembled a Stacking Ensemble model combining Logistic Regression, KNN, and Decision Trees, attaining 91.3% accuracy for heart disease detection, outperforming individual models.
Technology: Python Machine Learning Predictive Modeling Supervised Learning sklearn Pandas numpy streamlit SVM Random Forest XGBoost ADABoost Logistic Regression Feature Engineering SMOTE Evaluation Metrics AUC-ROC

Intelligent Food Management Application

Nov 2022

  • Created an intelligent food management app, leveraging Laravel and JavaScript, resulting in a 30% reduction in food waste.
  • Implemented an algorithm that calculated meal options from available ingredients, leading to a 25% reduction in grocery spending.
Technology: Laravel(PHP) HTML/CSS JavaScript Bootstrap MySQL Git GitHub REST API

Social Media platform for Researchers and Developers

Mar 2022

  • Developed a web application for programmers and researchers to efficiently collaborate and network.
  • Engineered a recommendation system connecting users with relevant projects and peers, leveraging skillsets and coding styles.
  • Formulated a data-driven user matching algorithm using weighted scoring to connect researchers and developers based on shared interests and skills, enhancing user experience and interdisciplinary collaboration.
  • Enhanced database performance by optimizing SQL queries and introducing indexed views in MySQL, which sped up data retrieval by 40% for complex searches.
  • Achieved Top 10 ranking among 120+ competing teams in a Hackathon, showcasing exceptional project execution and innovation.
Technology: Laravel(PHP) HTML/CSS JavaScript Bootstrap MySQL Git GitHub GitHub API

Intelligent Food Management Application

Nov 2022

  • Created an intelligent food management app, leveraging Laravel and JavaScript, resulting in a 30% reduction in food waste.
  • Implemented an algorithm that calculated meal options from available ingredients, leading to a 25% reduction in grocery spending.
Technology: Laravel(PHP) HTML/CSS JavaScript Bootstrap MySQL Git GitHub REST API

PUBLICATIONS

Research Patent: Enhancement of Advanced Encryption Standard Algorithm to secure IoT devices.

Aug 2022

  • Indian Patent Application No. 202221045319 A, The Patent Office Journal No. 33/2022, Date of filing: Aug 8, 2022
  • Proposed a faster alternative to the AES algorithm for lower-powered IoT devices and real-time secure communications

Research Project: Offensive Web Application Security Framework | ICETESM

Feb 2022

  • Assembled a sophisticated scanning engine to detect various web app vulnerabilities, utilizing open-source tools for in-depth security analysis.

Research Project: Enhancing Steering Accuracy in Self-Driving Cars Using Deep Learning | IJARESM

Feb 2022

  • Preprocessed 10,000+ images and addressed dataset bias by eliminating 30% of 0-degree steering angles, improving training efficiency by 15% and boosting performance on curves and complex road structures by 20%.
  • Devised a custom CNN with 5 convolutional, 4 dropout, and 4 dense layers, achieving training loss of 0.0343 and validation loss of 0.0275, enabling accurate steering predictions and improved generalization.
  • Minimized overfitting by applying 4 dropout layers in the CNN, reducing validation error by 25% and enhancing generalization.

LEADERSHIP / EXTRA-CURRICULAR ACTIVITIES

Technical Head at National Service Scheme (NSS) - TSEC, University of Mumbai

Sep 2021 - Jun 2022

- Organized over 80 community service events including Blood Donation Drives, Cleanup Drives, etc.

General Body Member at Rotaract Club - TSEC, University of Mumbai

Aug 2021 - May 2022

- Volunteered at various Medical camps, Fundraising, and charity events.

AWARDS & ACHIEVEMENTS

LEADERSHIP AWARD - TSEC, University of Mumbai

Mar 2023

Best Instrument Award

Dec 2020

3rd Place in Solo Singing Award

Dec 2020

1st Place in Solo Singing Award

Dec 2017

MY MUSIC

When 'Rolling Stone' - the world's biggest music magazine featured my song, it felt like a dream came true. Eight years of creating music and using software applications for music production have made me realize how evolving technology and Artificial Intelligence have transformed the way creativity can be expressed.