I'm Vemula Sai Saketh, a dynamic software engineer with a passion for building innovative solutions and delving deep into the intricate world of data. Currently based in Bloomington, IN, I am pursuing my Master's in Computer Science at Indiana University Bloomington.
I'm open to exciting opportunities, collaboration, and discussions on all things tech. Feel free to reach out for a chat!
Contributing to the System Programming course with C & Unix. Designing assignments in C language, implementing automated testing, and grading students' work in under 5 seconds using Shell scripts.
Developed 35+ Microservices as REST APIs (Flask, Fast API, Waitress) for user authentication, data insights posting, and tracking processes & threads status. Tested with Postman.
Built a Python-based web server framework for real-time processing of multiple Computer Vision ETL pipelines using Gstreamer, OpenVino, multithreading, and multiprocessing techniques.
Collaborated with DevOps to containerize the app in Docker, enhancing efficiency by 50% through streamlined CI/CD deployment on Bitbucket with automated testing and Docker Hub image updates.
Deployed a Linux service to monitor CPU, GPU, RAM, & app status in real-time. Utilized Lambda, API Gateway for data extraction, and integrated with React Dashboard. Stored data in DynamoDB.
Achieved a 25% system performance improvement by implementing Object-Oriented Programming System style. Source controlled the code on Bitbucket.
Designed and developed reusable UI components to enhance healthcare imaging solutions using AngularJS. Wrote Unit/Integration tests to improve interface reliability and reduced bug reports by 25%.
Designed, Built, and Implemented REST APIs in SharePoint portals using C# and queried SQL databases to meet the business and data needs of an organization.
I am a Master's student in Computer Science at Indiana University, where I have been performing exceptionally well, with a consistent grade point average of 4.0/4.0. I have completed a range of courses in the field, including Applied Algorithms, Cloud Computing, Data Mining, Applied Machine Learning, Computer Networks, Software Engineering, Applied Database Technologies, Security for Network Systems and Cyber Defense Competitions among others. This has provided me with a strong foundation in the field and a comprehensive understanding of the key concepts and technologies.
I completed my Undergrad studies in Computer Science at PES University, where I excelled academically with a CGPA of 8.36/10. My curriculum emphasized on the foundation of programming and computer science, and I took courses such as Object Oriented Modeling and Design, Data Structures, Big Data, Operating System, and Web Development I & II, to deepen my understanding of these core concepts.
Developed Dockerized Tensorflow face recognition app. Designed & implemented a CI/CD pipeline using Jenkins & Ansible, allowing seamless integration, automated builds of the latest versions, updates on DockerHub, & deployment on Kubernetes. Resulted in enhanced scalability & ensured high availability.
Led 5-member team in creating a Venue Management System Web App with NodeJS, ReactJS, and Redux. Integrated ExpressJS for 20+ REST APIs, used Nodemailer for email OTPs, performed CRUD on MongoDB, packaged in docker & deployed on Heroku.
Built ETL pipeline on AWS Lambda & Eventbridge for weekly extraction of 50k JSON records from Spotify API. Stored transformed & normalized data in DynamoDB. Loaded data on ReactJS & ChartJS dashboard via API Gateway & Lambda with set IAM permissions. Achieved under 10s processing time.
Developed US drought classification models using weather & soil time series data. Conducted data cleaning, EDA, visualization with Matplotlib, & utilized Standard Scaler, PCA, GridsearchCV for model optimization. Implemented Random Forest, Bagging Classifier, Naive Bayes, achieved 90.5% accuracy.
Trained a Machine Learning Classification (2 classes) model to predict loan repayment capability of customers applying for home credit. Evaluated accuracy using baseline logistic regression. Further, worked on feature engineering and hyperparameter tuning using GridSearchCV, Decision trees, KNN, Random Forest, XGBoost to increase AUC score and achieved accuracy of 92%.
Built a Docker-packaged face recognition SAAS application, utilizing Deepface (TensorFlow) for real-time detection from live video streams. Deployed the system using AWS CloudFormation & Fargate with auto-scaling for high availability, managed traffic distribution using a load balancer across multiple application instances in a round robin fashion.