● In my current role as a Data Analyst at LJ Magazacilik, I have developed and implemented a product transfer and stock optimization software that supports 24 stores. This system, built using Python,optimizes inventory management and has significantly enhanced operational efficiency. ● I conduct weekly sales performance analyses to facilitate data-driven decision-making for product transfers, leveraging tools such as Excel, Dataiku, and BigQuery alongside Python.

● Prepared datasets for testing and training algorithms, preparing datasets for image processingapplications. ● Prepared the dataset for training and test operations using the OpenCV library. ● Worked on machine learning algorithms for detecting roads on satellite images (UNet). ● Worked on platforms such as Kaggle, Roboflow, Google Colab, and Juptyer Notebook. Used Python API for online datasets. ● License Plate Recognition System (reading plates with Yolo format). Python, CVAT, Linux.

● Implemented image-processing applications using the OpenCV library. ● Worked on the image processing functions of the vehicle license plate recognition system realized bythe company and took part in the debugging processes.

● Took part in the Backend part for the GRC website the company is developing. ● Between October 2021 and June 2022, I worked on developing a new website and creating new add-onsto the existing mobile application as part of the graduation project.

● Cleaning datasets using Numpy and Pandas libraries. ● Using the BeatifulSoup library to extract appropriate information from websites (web-scrapping, web-crawling). ● Matching the information obtained by web-crawling with datasets. Projects: AB-Testing. ● AB testing (A/B testing or split testing) is a statistical experimentation process used to compare two or more versions and determine which version performs better. ● In this project, the datasets are handled in two different ways as control and test. After the control and test datasets are examined, they are brought together. AB hypotheses are defined and then Assumption Checks are performed. ● Shapiro test for normality assumption, levene test for variance homogeneity are created.In this project, two sample t test was applied if the assumptions are met. Association Rule Based Recommender System. ● Association Rule Based Recommender System can be defined as a type of recommender system that provides product recommendations based on users' past purchases or preferences. In this project, the quanity and price values are notimportant for us, and the suppression process is applied according to the z-scores of the variables. ● Rules are determined for the targeted customer groups, and then product recommendations are made to the target audience with the mlxtend library.