Machine Learning Data Science

Ayush Sojitra

+1-214-***-**** *************@*****.*** linkedin.com/in/ayush-sojitra github.com/AyushSojitra EDUCATION

The University of Texas - Dallas, Texas, USA May 2024 Master of Computer Science GPA: 3.93/4.0

Graduate coursework: Machine Learning, Database Design, Statistics for Data Science, Design and Analysis of Algorithm. Charotar University of Science and Technology, Gujarat, India May 2022 Bachelor of Technology in Computer Science and Engineering CGPA: 9.11/10.0 Courses: Artificial Intelligence, Machine Learning, Operating Systems, Data Structure, Analysis of Algorithm, Database, Blockchain. TECHNICAL KNOWLEDGE

Programming Languages: Python, C, C++, R, SQL, JavaScript Python & DL Libraries: NumPy, Pandas, Matplotlib, PyTorch, TensorFlow, NLTK, OpenCV, Sklearn, PySpark ML & DL Techniques: Regression, Clustering, SVM, Feature Engineering, Random Forest, NLP, CNN, GAN, YOLO Analytics & Other Frameworks: Tableau, Excel, PowerBi, Hadoop, Spark, DataBricks, ETL, AWS, Flask, Git PROFESSIONAL EXPERIENCE

Avea Media Labs, India - Data Science Intern December 2021- May 2022

• Implemented a cutting-edge parking system using YoloV4 object detection algorithm to achieve 96% real-time accuracy in vehicle detection, resulting in a seamless and efficient parking experience for users.

• Streamlined the functionality of the system by utilizing MQTT to transmit acknowledgements over the server after vehicle detection, improving overall communication.

Harikrushna Technologies, India - Machine Learning Engineer Intern May 2021- July 2021

• Developed and executed a comprehensive data collection plan by manually creating a dataset of 2500 paired images (with & w/o background) using Adobe Illustrator, resulting in highly accurate training data for deep learning models.

• Programmed a deep learning model for background removal using Pix2Pix and Cycle GAN to attain and accuracy of 94%.

• Expedited the company’s image processing tasks by 34% using the optimized deep learning model. Augmetic Infinite LLP, India - Software Development Intern April 2020- July 2020

• Collaborated with a team of 3 developers to successfully build and launch the "Prediction Arena" web application.

• Implemented seamless integration between the Frontend and SQL Database by utilizing JavaScript and PHP. PROJECTS

Sign Language to Text conversion (Python, Neural Networks, Image Processing, OpenCV) November 2021

• Engineered a CNN model trained on over 5000 images to perform real-time image classification to deduce ASL letters and words using gesture. Obtained an F1 score of 0.89 after hyperparameter tuning.

• Implemented a real-time ASL to text conversion application, processing 60 frames per second, resulting in improved communication for the deaf community.

Pneumonia Detection (Python, Tensorflow, Sklearn, Transfer Learning, Flask, JavaScript) August 2021

• Utilized transfer learning techniques and augmented data to train three highly accurate pneumonia detection models using layered CNN with VGG16, VGG19, and ResNet50 architectures, achieving an average accuracy of 98.5%.

• Developed a user-friendly web application using flask & JavaScript to allow doctors and healthcare professionals to easily upload chest X-ray images for quick and reliable diagnosis, resulting in a significant reduction of diagnosis time. Data Analysis using PowerBi (PowerBi, MySQL WorkBench, SQL, Analysis, Data Visualization) January 2021

• Conducted data cleaning on a sales database with over 100000 rows using SQL, DAX, and m to ensure accuracy and integrity.

• Developed an interactive dashboard using PowerBi by connecting MySQL database and performing ETL that provided insights into key performance metrics such as revenue growth and customer churn rate. Real Estate Price Prediction (Python, NumPy, Pandas, Linear Regression, Flask, AWS) November 2020

• Utilized the clean dataset obtained from extensive exploratory data analysis and feature engineering to create a robust Linear Regression Model, resulting in price predictions with an accuracy of 91%.

• Implemented a user-friendly web application using flask framework, allowing users to input property details and receive price predictions based on the developed Linear Regression Model.

• Streamlined the deployment process by hosting the web app on AWS EC2, configuring NGINX as a reverse proxy server for efficient handling of traffic. Additionally, utilized Putty and WinSCP for seamless remote management and file transfer. PUBLICATIONS

• Pneumonia Detection using Transfer Learning, International Journal of Engineering Research & Technology (IJERT)

• Accelerated Data Science using Rapids, International Journal of Research and Analytical Reviews (IJRAR) LEADERSHIP EXPERIENCE

Intramural Sports Lead, The University of Texas at Dallas October 2022 - Present

• Oversee and Co-ordinate the management of university-level sports leagues.

• Supervise and collaborate with the referees officiating the games. IEEE Depstar Student Branch, Charotar University of Science and Technology February 2020 – December 2020

• Organized and facilitated 15+ workshops, competitions, and expert talks within the organization while collaborating effectively with team members.

Contact this candidate