Anteon is an open-core, high-performance distributed observability platform focusing on Kubernetes (K8s) monitoring using eBPF and performance testing.GitHub: https://github.com/ddosify/ddosify

► Promoted to 𝗧𝗲𝗮𝗺 𝗟𝗲𝗮𝗱 of the Computer Vision team, leading a group of four Data Scientists.► Collaborated with various stakeholders, including 𝗔𝗰𝗮𝗱𝗲𝗺𝘆 𝗮𝗱𝘃𝗶𝘀𝗼𝗿𝘀, 𝗜𝗧, 𝗙𝗿𝗼𝗻𝘁𝗲𝗻𝗱, and 𝗕𝗶𝗴 𝗗𝗮𝘁𝗮 𝗧𝗲𝗮𝗺.► Led the development and successful release of high-accuracy, real-time custom vision models.► Managed the entire model development lifecycle from initial conception to 𝘁𝗿𝗮𝗶𝗻𝗶𝗻𝗴, 𝗱𝗲𝗽𝗹𝗼𝘆𝗺𝗲𝗻𝘁 𝗮𝗻𝗱 𝗶𝗻𝗳𝗲𝗿𝗲𝗻𝗰𝗲.► Created training and inference environments using 𝗞𝘂𝗯𝗲𝗿𝗻𝗲𝘁𝗲𝘀, 𝗞𝗮𝗳𝗸𝗮, 𝗥𝗲𝗱𝗶𝘀, 𝗥𝗧𝗦𝗣, 𝗴𝗥𝗣𝗖, and other technologies in an 𝗮𝗶𝗿-𝗴𝗮𝗽𝗽𝗲𝗱, multi-cluster Linux environment.► Executed extensive testing of models, including 𝘃𝗮𝗹𝗶𝗱𝗮𝘁𝗶𝗼𝗻 and 𝗹𝗼𝗮𝗱 𝘁𝗲𝘀𝘁𝗶𝗻𝗴, to ensure reliability and efficiency.

► Prepared and cleaned extensive AI datasets, particularly for Computer Vision, which is crucial for training deep learning models► Checked the annotated data quality (100+ hours of videos) and relevance to achieve optimal model performance► Trained deep learning models for 𝗢𝗯𝗷𝗲𝗰𝘁 𝗗𝗲𝘁𝗲𝗰𝘁𝗶𝗼𝗻, 𝗙𝗮𝗰𝗲 𝗗𝗲𝘁𝗲𝗰𝘁𝗶𝗼𝗻, and 𝗙𝗮𝗰𝗲 𝗥𝗲𝗰𝗼𝗴𝗻𝗶𝘁𝗶𝗼𝗻► Trained the models on multi-GPU clusters on the 𝗡𝗩𝗜𝗗𝗜𝗔 𝗗𝗚𝗫-𝟭 𝗣𝗹𝗮𝘁𝗳𝗼𝗿𝗺 (𝗩𝟭𝟬𝟬) with 𝗞𝘂𝗯𝗲𝗿𝗻𝗲𝘁𝗲𝘀 to accelerate the process► Utilized CI/CD, Test, and 𝗠𝗟𝗢𝗽𝘀 practices with tools such as Weights & Biases and neptune.ai► Continuously refined models through iterative training, achieving significant improvements in accuracy and reliability► Successfully deployed AI models in real-world scenarios for real-time inference using technologies like 𝗞𝗮𝗳𝗸𝗮, 𝗥𝗧𝗦𝗣, and 𝗡𝗩𝗜𝗗𝗜𝗔 𝗧𝗲𝗻𝘀𝗼𝗿𝗥𝗧.

► Implemented 𝗖𝗜/𝗖𝗗 pipelines using 𝗝𝗲𝗻𝗸𝗶𝗻𝘀 to automate the software development.► Automated manual unit, integration, and UI tests utilizing 𝗦𝗲𝗹𝗲𝗻𝗶𝘂𝗺, 𝗔𝗽𝗽𝗶𝘂𝗺, and other technologies.► Utilized 𝗗𝗼𝗰𝗸𝗲𝗿 and 𝗞𝘂𝗯𝗲𝗿𝗻𝗲𝘁𝗲𝘀 for efficient and scalable deployments.► Deployed monitoring solutions like 𝗚𝗿𝗮𝗳𝗮𝗻𝗮 and 𝗣𝗿𝗼𝗺𝗲𝘁𝗵𝗲𝘂𝘀.► Worked extensively with 𝗢𝗿𝗮𝗰𝗹𝗲 𝗗𝗮𝘁𝗮𝗯𝗮𝘀𝗲 and 𝗝𝗮𝘃𝗮 programming language, enhancing backend functionalities.► Managed applications on 𝗢𝗿𝗮𝗰𝗹𝗲 𝗪𝗲𝗯𝗟𝗼𝗴𝗶𝗰 and 𝗧𝗼𝗺𝗰𝗮𝘁 servers, ensuring robust and reliable service performance.► Contributed to the Work Force Management (WFM) system project, focusing on improving operational efficiency.► Actively involved in the Telco/BSS - OSS department, addressing essential telecommunications industry requirements.

► Worked as a 𝗠𝗮𝗰𝗵𝗶𝗻𝗲 𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴 𝗜𝗻𝘁𝗲𝗿𝗻 in the Information and Security Technologies, Image Processing Unit.► Developed deep learning applications using 𝗠𝗔𝗧𝗟𝗔𝗕 and the 𝗖𝗮𝗳𝗳𝗲 framework.► Contributed to a project focused on 𝗹𝗶𝗰𝗲𝗻𝘀𝗲 𝗽𝗹𝗮𝘁𝗲 𝗿𝗲𝗰𝗼𝗴𝗻𝗶𝘁𝗶𝗼𝗻 and 𝘃𝗲𝗵𝗶𝗰𝗹𝗲 𝗺𝗼𝗱𝗲𝗹 𝗽𝗿𝗲𝗱𝗶𝗰𝘁𝗶𝗼𝗻.► Used 𝗗𝗲𝗲𝗽 𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴 techniques for detecting cars as a part of the project implementation.

► Worked as part of a team on a 𝗩𝗶𝗿𝘁𝘂𝗮𝗹 𝗥𝗲𝗮𝗹𝗶𝘁𝘆 (𝗩𝗥) project► Used 𝗨𝗻𝗶𝘁𝘆𝟯𝗗, 𝗞𝗶𝗻𝗲𝗰𝘁, and 𝗟𝗲𝗮𝗽 𝗠𝗼𝘁𝗶𝗼𝗻► Wrote software for the Zeugma Museum, with a focus on providing indoor navigation service through beacons► Managed and controlled the museum's software systems on the server side► Used 𝗝𝗮𝘃𝗮 𝗣𝗹𝗮𝘆 𝗳𝗿𝗮𝗺𝗲𝘄𝗼𝗿𝗸 for server control and management tasks.

► Intern at Zorlu Holding, Department of Software Engineering► Developed the backend and a basic frontend side of the website for students who receive scholarships from the Mehmet Zorlu Foundation► Used technologies including 𝗠𝗦𝗦𝗤𝗟 𝟮𝟬𝟭𝟰, 𝗖#, 𝗕𝗼𝗼𝘁𝘀𝘁𝗿𝗮𝗽, 𝗖𝗦𝗦, 𝗛𝗧𝗠𝗟, 𝗘𝗻𝘁𝗶𝘁𝘆 𝗙𝗿𝗮𝗺𝗲𝘄𝗼𝗿𝗸, 𝗷𝗤𝘂𝗲𝗿𝘆, and 𝗔𝗷𝗮𝘅 for web development► Integrated 𝗪𝗲𝗯 𝗦𝗲𝗿𝘃𝗶𝗰𝗲𝘀 to enhance the functionality and user experience of the site► Involved in the entire project lifecycle, from initial design to implementation and deployment► The project can be viewed at: https://bursiyer.mzv.org.tr/holding/login