Introduction: Defining the Problem In today’s digital world, misinformation and fake news spread faster than ever. To tackle this problem, I decided to build a machine learning model that can analyze the credibility of text-based content. I named this project “CheckMate” — it includes a deep learning model built from scratch to detect fake news and exposes the model via an API. Project Goal and Scope The core goal of CheckMate is to classify a given text input as “real” or “fake” with high accuracy. The main objectives were: ...

As a team leader and full-stack developer in this project, I developed an AI-powered automation system that manages processes such as order management, menu suggestions, and customer interaction. Python and SQLite were used for backend and data management, while the user interface was created with HTML, CSS, and JavaScript.

I designed and developed a smart recycling system that automatically sorts waste by type (glass, paper, plastic, etc.) using image processing technology. In this project, I used the YOLO algorithm along with Python and OpenCV libraries for object detection. This project won the first prize in the Teknofest Idea Competition.

I developed a modular Red Team tool with a command-line interface (CLI) to automate cybersecurity operations. This Python-based tool was designed as a framework housing various modules such as network discovery, service scanning, and privilege escalation in Linux systems.

As a team leader, I managed the project to increase the stability and debug the Pardus operating system. Within this scope, we performed open-source code analysis, reviewed system logs, and conducted debugging processes to identify potential errors. The project provided in-depth practical experience in Linux system administration and open-source development methodologies.

Based on a real-world scenario, I intervened in a multi-stage cloud data breach incident. In the AWS environment, I analyzed S3 data event logs (s3_data_events) and CloudTrail logs using SQL queries. As a result of this analysis, I identified that the attacker had compromised the S3Reader role with STS AssumeRole, the IAM user Moe.Jito who performed this action, and the Lambda function named credrotator whose code was manipulated for exfiltration. Deepening the investigation, I found the source IP address from the SSH logs on the EC2 server, which was the attacker’s initial access point. In the final stage, I connected to an external HTTP server with curl to delete the confidential data before it was disclosed and prepared a comprehensive incident response report including all findings. ...

I developed an object detection system capable of real-time detection of humans, vehicles, and various objects. In this project, I integrated the YOLOv8 model using Python and OpenCV libraries and gained practical experience in image processing applications.