Oleksandr Petsa

Hello! I am working on projects related to artificial intelligence (AI) and machine learning. I write in Python and create scripts. I am constantly improving my skills and am open to interesting projects and collaboration.

Recognition of generated photos: 

• This project implements a deep learning model for recognition real and generated images using TensorFlow/Keras, structured with modular components and configurable path. 

Bachelor's work: 

• This bachelor's work is about analyzing data and setting up monitoring screens using the Grafana visualization platform. Its main goal is to develop and implement a data visualization program for effective monitoring of processes and systems. The work focuses on analyzing available data sources, connecting them to the PostgreSQL database system, and then processing the data for visualization. The proposed methodology includes data collection, transformation, and storage, monitoring panel design, and Grafana configuration to display key parameters. The result is the creation of a monitoring solution that allows visualization of important indicators and supports the decision-making process in system management. The last part of the thesis evaluates the functionality of the implemented solution and its contribution to effective monitoring. 

AITS: 

• My task was to visualize data within Google Cloud Monitoring, collect metrics from a virtual machine using Prometheus, and transfer them to Grafana for real-time dashboarding, alerting, and performance analysis of the system. 

• I also recently finished working on a logistics task that focused on finding the shortest route between points. It works as follows: a json file is sent to a server running on FAST API, then Google Maps API (Routes API) is used, which shows us the distance from each point to every other point and returns a matrix of all distances. Then we use the TSP algorithm to select the shortest path — first, the closest next point is selected from the starting point, then the closest point from that closest point, and so on. At the end, a route is also built from the last point to the starting point, i.e., we have a closed circle. The project is on my GitHub.