Data science

I have 10+ years of experience processing data in high volumes and developing data science solutions within MANGO, HP and the ATLAS Collaboration

Top skills, software & tools

Statistics and Machine Learning: PyTorch, Scikit-learn, TensorFlow, XGBoost, SciPy & statsmodel
Advanced Data Analysis: SQL & Python (PySpark, Pandas & NumPy)
Data Visualization and Data Analytics: Tableau, Looker Studio, Plotly & Matplotlib
Databricks, Jenkins, Apache Airflow, Docker, Kubernetes [Kubeflow]
Git [GitLab and GitHub], Continuous Integration and automated testing (Pytest)
Jupyter notebook
Microsoft Excel and Google Sheets

End-to-end development of Machine Learning solutions to optimize purchases
- Tech stack: python, PySpark, PyTorch, sklearn, Databricks, Apache Airflow, Jenkins and GitHub Copilot
Extraction of data-driven insights enabling decision making
Active participation in the interview process for new candidates, contributing to the selection of talented professionals who align with our team's goals

End-to-end development of Machine Learning solutions to increase sales across business units (HP Store and Channel), including propensity-to-buy and revenue prediction models
Built an end-to-end Machine Learning model that recommended PCs to clients, driving sales and revenue while adhering to business rules and constraints
Performed A/B testing with power analysis, sample size estimation, and test/control strategy
Drove SMB account–agent assignment strategy with actionable recommendations, supported by dashboards, economic impact analysis, and data-driven insights to guide agent conversations
Developed and maintained data pipelines for data science workflows, implementing fixes, tests, and automation, and designing a data quality framework for continuous data validation.
Active participation in the interview process for new candidates, contributing to the selection of talented professionals who align with our team's goals
Tech stack: python, PySpark, pandas, SQL, sklearn and xgboost

Performed several data analyses comprising:
- Data preparation and cleaning
- Precision measurements of physical quantities
- Various statistical analyses:
  - Extraction of data-driven corrections
  - Determination of uncertainties
  - Test statistic based on profile likelihood ratio for hypothesis testing
  - Chi-square goodness of fit test for hypothesis testing
- Successfully edited and published 5 scientific results
Implementation and deployment of a neural network to predict the position of a particle in a detector:
- Improved previous estimation by up to 60%
- Using NumPy, Pandas and Keras from tensorflow
Training and optimized an attention-based model to identify physical particles:
- Improved performance by up to 50%
- Using Docker and Kubernetes (Kubeflow pipelines and Katib)

Project Manager

Software Developer

Physicist