Fariborz Bagherzadeh

• Designed and deployed end-to-end Python/SQL ML pipelines across 3+ building systems and 5+ data sources, covering ingestion, feature engineering, training, evaluation, and monitoring. • Built and benchmarked LSTM, CNN, and regression models with experiment tracking, achieving RMSE improvements of 15–20% over baselines on 18+ months of time series data. • Implemented production-style data quality and drift monitoring on live streams, detecting 100% of distribution shifts before model updates. • Created

• Led end-to-end machine learning projects, including data acquisition, preprocessing, modeling, evaluation, and visualization. • Applied supervised and unsupervised techniques for forecasting, classification, and clustering tasks. • Conducted in-depth data analysis on large datasets, extracting actionable insights for research purposes. • Mentored peers in Python, machine learning algorithms, and statistical methods to enhance collaborative learning.

Master thesis: Comparative evaluation of drift detection methods for incremental fine-tuning in heat load forecasting