Portfolio

Summary:

Motivated Software Developer with a Master of Science in Computer Science from Portland State University. Proficient in a wide range of programming languages including Python, C#, and Java, with expertise in developing innovative solutions using machine learning, computer vision, and game development technologies. Strong background in project development, mentorship, and collaboration.


Technical Skills:

  • Languages: Python, C#, HTML, CSS, PHP, JavaScript, Java.
  • Databases: MySQL, PostgreSQL, MongoDB.
  • Tools and Frameworks: Android Studio, Xcode, GitHub, Git, GitLab, Laravel, Unity - Game Engine, Tensorflow, PyTorch.

Projects:

architecture of epoch wise performance

  • Developed a depth layer between hidden layers within a neural network architecture for a linear regression problem in Python, resulting in an overall improvement of 100 times in accuracy for the test set per epoch.

coexistence evaluation graph

  • A ChatGPT (LLM) application addressing personal, social, and interpersonal challenges, aligning with the Coexistence Philosophy, resulting in 50+ conversations in 9 days.
  • Compiled and evaluated a 21-question dataset on 9 topics with 3 difficulty levels, employing 8 criteria, including relevance and critical thinking, to assess response quality.

sample images of positions

  • Generated an extensive dataset comprising 8,000 images to facilitate the localization of three-dimensional objects, utilizing C# in conjunction with the Unity - game engine.
  • Developed and assessed a Convolutional Neural Network model using TensorFlow in Python, achieving 95% accuracy on image classification tasks with 64x64 pixel input dimensions.
  • Developed and trained a generative subnetwork model in Python using TensorFlow to generate synthetic images, tailored for the FashionMNIST dataset containing 60,000 images with dimensions of 28x28 pixels.
  • Enhanced the model's capabilities through the implementation of a CNN-based GAN, resulting in the generation of high-quality, clear, and recognizable images with minimal noise.
  • Achieved grade A for the associated coursework.

sentiment detection confusion matrix

  • Implemented machine learning models in Python (Naive Bayes and Logistic Regression) for sarcasm detection in 26,709 article headlines.
  • Achieved 80% accuracy with Logistic Regression, a 9% improvement over Naive Bayes and grade A for the associated coursework.

sentiment detection confusion matrix

  • Developed a 3d puzzle game where intersection gives opposite color to figure out destination shape in C# and Unity - Game engine, resulting in 100 plus downloads.

Education:

Master of Science, Computer Science (March 2022 – December 2023) Portland State University, Portland, USA. GPA : 3.82

Bachlor of Engineering, Computer Engineering (July 2013 – June 2017)
Gujarat Technological University, Gujarat, India.


Previous Projects:

Abscond - 3rd Person Shooting Game

  • Developed third person controller mechanism, input controller, 3d models and A star search algorithm in enemies AI to catch the detected player in C# using Unity - game engine, resulting in project selection for yearly state level project demonstration.

World Speed Shopping - E-commerce Site (Laravel, PHP, HTML, CSS)

  • Developed Laravel-based e-commerce platform with secure authentication and PayUMoney integration, facilitating secure and efficient online transactions for physical goods.

Tablegroups - Social Network (Laravel, PHP, JavaScript, HTML, CSS)

  • Designed and developed Laravel-based social network with authentication, notifications, and relationship modules, fostering community engagement among students.

E-doctor - Healthcare Application (Java, Android Studio)

  • Designed and implemented Java-based healthcare application with secure sign-up, login, recommendation, and search functionalities, providing doctors and patients with a convenient platform for accessing healthcare services.

Farm ERP Web Application (Yii2.0, PHP, HTML, CSS, JavaScript)

  • Developed Yii2.0 PHP-based ERP system with modules for livestock management and natural activity cycles, enhancing farm and dairy management capabilities for efficient operations.

Feel free to reach out to discuss collaboration opportunities or for further details on any projects or experiences. Looking forward to connecting!