Oleksandr

Oleksandr Dmytrenko
Junior Python Developer | Software Enginee r
Würzburg, Germany | [open contact info](look above in the "contact info" section) | [open contact info](look above in the "contact info" section)
GitHub: github.com/Shershl | LinkedIn: [open contact info](look above in the "contact info" section)
SUMMARY
Junior Python Developer with hands-on experience building Python applications, REST APIs, data-processing
tools, and machine-learning services. Developed end-to-end projects using Python, FastAPI, pytest, Git,
PostgreSQL, PyTorch, and OpenCV, with a focus on modular code, API development, testing, data preparation,
and model inference. Bachelor’s degree in Automation and Computer-Integrated Technologies.
TECHNICAL SKILLS
Programming: Python
Backend & APIs: FastAPI, REST APIs
Testing & Development: pytest, Git
Data & Databases: PostgreSQL, NumPy, pandas
Computer Vision & ML: PyTorch, YOLOv8, OpenCV
Additional: Dataset Processing, File Processing, SHA-256 Hashing
PROJECTS
Traffic Object Detection API — Python / Computer Vision Project | 03/2026
Built an end-to-end object detection service using Python, FastAPI, YOLOv8, and OpenCV.
Developed a preprocessing pipeline to convert VisDrone annotations into YOLO format and map source categories
into five target classes.
Fine-tuned a YOLOv8 model and implemented an inference pipeline returning object classes, confidence scores,
bounding boxes, and annotated prediction images.
Developed REST API endpoints for image upload, object detection, health checks, and serving annotated
prediction images.
Structured the application into separate data-processing, training, inference, and API modules.
AI Dataset Inspector — Python CLI Project
Developed a command-line tool in Python for analyzing and validating image datasets.
Implemented recursive file scanning and validation for missing, empty, and unsupported image files.
Implemented duplicate-file detection using SHA-256 hashing.
Created automated unit tests with pytest for file validation, hashing, duplicate detection, and directory scanning.
Conditional Self-Attention GAN — Deep Learning Project
Implemented conditional Generator and Discriminator architectures in PyTorch for satellite image generation
across four classes.
Implemented custom self-attention layers and integrated spectral normalization into the discriminator.
Built a custom dataset pipeline and trained the model using Hinge Loss, Adam optimization, and mixed-precision
training.
Implemented checkpointing, training-loss logging, and generated-sample visualization.
EDUCATION
Bachelor’s Degree — Automation and Computer-Integrated Technologies | 2021 – 2025
Kyiv Aviation Institute, Kyiv, Ukraine
Exchange Student — Mathematics | 2022 – 2023
Julius-Maximilians-Universität Würzburg, Würzburg, Germany
DAAD Scholarship
CERTIFICATION
CS50P: CS50’s Introduction to Programming with Python — HarvardX / edX, 2026 | Verified Certificate
LANGUAGES
English: B2 | German: B1; B2 in progress | Ukrainian: Native | Russian: Native