Vladyslav

Vladyslav Haslo
[открыть контакты](см. выше в блоке «контактная информация») | [открыть контакты](см. выше в блоке «контактная информация») | https://vilardcool.github.io/Portfolio

Technical Skills
Languages: C/C++, Python, Java, SQL, JavaScript, HTML/CSS, AMPL
Frameworks: React, Node.js, Express.js, Socket.IO
Developer Tools: Git, Google Cloud, Microsoft Azure, Render

Projects
Vilard’s Quiz | JavaScript, Node.js, Express.js, Socket.IO, Render July 2025 – Present
• Developed a full-stack web application with more than 100 users by implementing an interactive quiz platform with

real-time question rendering, scoring, and timer-based assessments
• Accelerated quiz creation workflow by 75% by building a packs creation tool with visualized main information

about questions and content
• Reduced quiz evaluation time by 90% by implementing automatic grading and instant result generation

• Enhanced user retention by 30% through responsive UI design and timer-based interactive quizzes

Vilard’s Image Platform | Python, JavaScript, AMPL, Keras, Render Sep. 2023 – May 2025
• Reduced image storage requirements by up to 92% by developing a lossy image compression platform using

aglomerative, optimisation and deep learning–based algorithms
• Accelerated compression processing time by 35% by creating aglomerative clustering image compression algorithms

• Improved the search for the most efficient representative color by 86% by implementing optimisation P-median

problem with AMPL
• Implemented a super-resolution algorithm that can significantly enhanced image quality of 586Ö391 pixel picture

while increasing file size by only 29 KB by training convolutional neural network with Keras

Telegram ChatBot | Python, Telegram.ext, PARCS, Google Cloud, Microsoft Azure Jan. 2022 – June 2023
• Reduced image transmission size by more than 88% by developing a Telegram chatbot implementing lossy image

compression using uniform quantization, K-means, vector quantization, median-slice, and octree-based methods
• Improved compression efficiency by 60% by integrating adaptive quantization strategies and clustering-based

encoding for optimal pixel representation
• Increased chatbot usability and response speed by designing lightweight compression workflows suitable for

Telegram API constraints and real-time messaging
• Enhanced system robustness by handling diverse image types through adaptive preprocessing and tunable

quantization levels based on image complexity
• Implemented parallel computing for K-means method, which showed the fastest result with low visual loss, that

requires up to 97% less time for processing by using the PARCS technology, Google Cloud and Microsoft Azure

Education
Taras Shevchenko National University of Kyiv Kyiv, Ukraine
PhD in Applied Mathematics Oct. 2025 – Present
Taras Shevchenko National University of Kyiv Kyiv, Ukraine
Master of Science in Computer Science with honors Oct. 2023 – June 2025
Taras Shevchenko National University of Kyiv Kyiv, Ukraine
Bachelor of Science in Computer Science Sep. 2019 – June 2023