Emre Sevik

Nice to meet you,

I am Emre

I am a 4th-year Computer Engineering student at Istanbul Aydın University with strong knowledge in machine learning, deep learning, and computer vision, and a strong interest in Artificial Intelligence Engineering. Proficient in SQL, HTML/CSS, and Java, with hands-on experience in data analysis, AI modeling, and web development. Actively use libraries such as NumPy, Pandas, Scikit-Learn, TensorFlow, and YOLO, while staying up to date with the latest AI and data science advancements. Additionally, I have knowledge of AI Agents architectures and automation tools like n8n, enhancing my ability to build intelligent systems.

View Projects Contact Me
Emre Sevik

Experience

AI Intern

Ithinka · Aug 2025 - Sep 2025
  • Contributed to YOLOv8-based data preparation and model training processes; performed image labeling (CVAT, Roboflow) and worked on training parameter optimization.
  • Developed a Flask-based web application for a face recognition attendance system; stored user face embeddings in pickle files and integrated them with a PostgreSQL database.
  • Implemented real-time entry/exit detection, face bounding boxes with names and confidence scores, and a dashboard to track employees' working durations.
  • Deployed the project on Heroku and optimized performance by replacing YOLO with the Face Recognition face_locations function to prevent overload issues.

AI & Computer Vision Intern

Emayer · Jul 2025 - Aug 2025
  • Developed and deployed real-time object detection pipelines using YOLOv8/v5-OBB, achieving up to 90% mAP in industrial automation tasks. Gained hands-on experience in database management and web development.
  • Implemented and evaluated advanced tracking algorithms (ByteTrack, OC-SORT) for robust video-based object tracking and classification tasks.
  • Conducted in-depth analysis of model performance metrics (precision, recall, mAP) and performed comparative tests between different YOLO versions and training strategies.
  • Gained hands-on experience in deploying AI models on edge devices (e.g., NVIDIA Jetson), optimizing for speed and accuracy in production environments.
  • Strengthened technical proficiency in Python, PyTorch, OpenCV, and MLOps best practices through project-based learning.

Education

ISTANBUL AYDIN UNIVERSITY

Bachelor of Science in Computer Engineering (English) · Sep 2022 - Present
  • Relevant Coursework: Machine Learning, Deep Learning, Data Structures, Algorithms, Database Management Systems, Artificial Intelligence

ISTANBUL AYDIN UNIVERSITY

English Preparatory School · Sep 2021 - Jun 2022
  • Relevant Coursework: Speaking, Writing, Reading Comprehension, Grammar

Skills

Programming Languages

Python, Java, SQL, JavaScript, HTML/CSS, PHP

Machine Learning & AI

NumPy, Pandas, Scikit-Learn, TensorFlow, YOLO, OpenCV, Data Preprocessing, Model Training & Evaluation

Computer Vision

YOLOv8/v5-OBB, Object Detection, Image Processing, Face Recognition, ByteTrack, OC-SORT

Web Development

Flask, RESTful APIs, Frontend-Backend Integration, Deployment, PostgreSQL, Heroku

Data Analysis & Visualization

Matplotlib, Seaborn, Streamlit, Data Analysis, AI Modeling

Certifications

  • Miuul Machine Learning Camp
  • Advanced Learning Algorithms – DeepLearning.AI
  • Supervised Machine Learning: Regression and Classification – DeepLearning.AI
  • Machine Learning Algorithms – BUSIBER Bogazici

Few of my projects

Cinematch - Movie Recommendation System

Designed and implemented a Django-based movie recommendation system featuring user authentication, personalized suggestions, favorites lists, and browsing of popular/top-rated movies. Developed a responsive web interface using HTML, CSS, and JavaScript to support seamless user interaction. Employed machine learning algorithms for recommendation logic, integrating Python-based models within the Django backend. Utilized MySQL for persistent storage of user data and movie-related records.

PET Bottle Sorting System (Emayer, 2025)

Contributed to the development of an industrial PET bottle sorting machine, utilizing YOLOv8 architecture for real-time object detection and classification. Integrated a conveyor system with advanced computer vision algorithms for automated and efficient separation of recyclable PET bottles. Achieved over 85% detection accuracy in live production tests, significantly increasing sorting efficiency and reducing manual labor. Actively involved in dataset creation, model training, and system validation as part of a multidisciplinary team.

Face Recognition Attendance System

Developed a Flask-based web application for a face recognition attendance system during internship at Ithinka. Stored user face embeddings in pickle files and integrated them with a PostgreSQL database. Implemented real-time entry/exit detection, face bounding boxes with names and confidence scores, and a dashboard to track employees' working durations. Deployed the project on Heroku with performance optimizations.

Click here to learn more projects

Let's build something great together

Open to collaborations, internships, and freelance work.

Email Me