Baiysh Narkeev

Developed an end-to-end machine learning solution using a structured pipeline. Leveraged Pandas for data cleaning and preprocessing, handling missing values, and performing feature engineering. Utilized Numpy for efficient numerical computations. Built predictive models with Scikit-learn, optimizing them through hyperparameter tuning and evaluating their performance using metrics like accuracy, precision, recall, and F1-score. Achieved 71% accuracy.

Visualized data insights and model performance using Matplotlib to create intuitive plots and charts. Integrated the solution into a FastAPI framework. Implemented version control using Git, ensuring collaborative development and code organization. Enabled robust experiment tracking and reproducibility with MLflow, logging metrics, parameters, and artifacts throughout the model lifecycle.

This project demonstrated my ability to handle the full ML pipeline, from data ingestion to deployment, with a focus on scalability, collaboration, and reliability.

Implemented a movie review classification project using transfer learning. Leveraged a pre-trained TensorFlow model to extract meaningful representations of text data. Evaluated model performance using classification metrics such as accuracy, precision, recall, and F1-score Visualized results and data insights with Matplotlib, and Achieved 85% Accuracy. This project highlighted the use of transfer learning to adapt pre-trained models for a specific text classification task