Sergiu Megiesan

Background

About Sergiu Megiesan

Data Scientist, Imperial College London Alumni - MSc in Applied Mathematics, Babes Bolyai University Alumni - BSc in Mathematics. As a passionate and results-driven Data Scientist, I specialize in leveraging advanced machine learning techniques and data analysis to deliver impactful solutions. My academic background in Applied Mathematics and hands-on experience have honed my expertise in solving complex problems and building scalable models. I am highly skilled in Python (including libraries such as NumPy, Pandas, SciPy, Scikit-Learn, TensorFlow, gpflow, XGBoost, hyperopt, TQDM, Keras, kmodes, Gower, Seaborn, and Matplotlib), MATLAB, SQL, Power BI, and AWS services (SageMaker, S3, Lambda, CloudWatch, Redshift). This technical toolkit empowers me to tackle all aspects of the data science pipeline, including- Data Cleaning, Exploratory Data Analysis (EDA), and Data Visualization - Feature Engineering and Dimensionality Reduction (e.g, Principal Component Analysis)- Predictive Modeling (Regression & Classification: Gaussian Processes, Decision Trees like XGBoost and Random Forest)- Clustering (e.g, K-Means, Dbscan, K-Prototypes, Gower Distance) and Segmentation - Model Selection and Optimization (e.g, Hyperparameter Tuning via Bayesian Optimization and Grid Search)- Building and deploying machine learning solutions on AWS platforms- Designed and executed A/B tests for multiple machine learning models to assess impact and improve accuracy. With experience spanning diverse industries, including automotive, banking and gambling, I have developed solutions such as customer churn prediction, product recommendation systems, player segmentation models, and early detection frameworks. My adaptability, problem-solving mindset, and eagerness to embrace new challenges ensure I deliver high-value insights and innovations in any project.