Hi, I'm Evint Leovonzko

Data Scientist building practical machine learning solutions

Explore My Work Get in Touch
ml_model.py
# Building the future with data
import tensorflow as tf
import pandas as pd

def solve_problems():
    return "innovative_solutions"

About Me

I'm a data scientist focused on developing machine learning solutions that solve real-world problems. With a background in computational research and business analytics, I bridge the gap between technical implementation and practical application.

My expertise includes deep learning algorithms, genetic algorithms, and building data analytics platforms that provide actionable insights.

Technical Skills

Languages & Frameworks

Python TensorFlow PyTorch Pandas NumPy

Visualization & Tools

Seaborn Matplotlib TensorRT CUDA

Specializations

Machine Learning Deep Learning Genetic Algorithms AutoML

Projects

Research

Deep Learning for Circular RNA Classification

Implemented a custom convolutional neural network architecture with Gaussian Blur data preprocessing to predict circRNA-disease associations. Achieved 75.11% accuracy, improving upon previous methods.

Python PyTorch NumPy Pandas
View Research →

Cancer Mutation Analysis in P-53 Gene

Identified specific DNA strand characteristics within the P-53 tumor suppressor gene that may lead to cancerous mutations, using genetic algorithms in a controlled environment model.

Python NumPy Genetic Algorithms Matplotlib
View Research →

Business

Dataktif

Built a data analytics engine that processes large-scale business data in real-time. Implemented machine learning models that convert complex data patterns into actionable business insights.

Python TensorFlow Pandas Matplotlib
Visit Platform →

Genta Technology

Developed solutions for AI workflows, including the Genta Inference Engine that optimizes model execution for production environments.

PyTorch TensorRT CUDA Python
Learn More →

Kolosal AI

Created an open-source desktop application for training and running large language models with optimized inference capabilities.

Python PyTorch CUDA TensorRT
View Project →

Educational Platform

rebirdth.id

Created an educational platform that empowers individuals to become knowledgeable in research methodology and fosters a community of lifelong learners. Developed tools and resources that provide practical tips, critical analysis of literature, and thought-provoking discussions to stimulate intellectual curiosity.

Python Machine Learning Data Analysis Web Development
Visit Platform →

Personal

Self-Organizing Map Algorithm

Implementation for unsupervised learning and clustering inspired by modern SOM techniques, with applications in data visualization and pattern recognition.

Python NumPy Matplotlib Seaborn
Project Details →

Particle Identification System

System using G4, genetic algorithms, and classification methods based on data importance for identifying particles in physics simulations.

Python TensorFlow NumPy Genetic Algorithms
Project Details →

Nuclear Accident Classification

Integration of machine learning and deep learning to detect and classify nuclear power plant incidents from operational data patterns.

Python PyTorch Pandas Seaborn
Project Details →

Kolosal AutoML

Integrated Machine Learning and CPU optimized platform to train and inference ML Model

Python Scikit Learn XGBoost Intel Optimization MLOps
Project Details →

Data Analysis

GDP Growth Analysis

Analyze various pattern of GDP growth for multiple countries and groups using SOM Clustering

Python NumPy Matplotlib Seaborn
Project Details →

Contact Me

Interested in collaboration or have questions? Let's connect:

LinkedIn Instagram GitHub