I’m an MSBA student focused on turning data into decision-making insights, especially in operations analytics. I like building reproducible ML workflows and writing clear notebooks/reports (EDA → modeling → validation → insights).

• 🔍 Interests: forecasting, risk/cost prediction, model validation, feature engineering
• 🧰 Tools: Python (pandas/numpy/sklearn), SQL, Tableau
• ✅ Goal: build a strong analytics portfolio through Kaggle + GitHub

• Competition: https://www.kaggle.com/competitions/mitsui-commodity-prediction-challenge
• Kaggle Notebook: https://www.kaggle.com/code/darcy59/mitsuico-single-lag-pair-group-lag-extratree
• GitHub Repo: https://github.com/Darcy59/kaggle-mitsui-commodity
  Highlights
• Single-lag + group-lag workflow to organize multi-target modeling
• ExtraTrees ensemble baseline for robust non-linear performance
• Clean Kaggle structure (stable paths + fixed seed)

• Competition: https://www.kaggle.com/competitions/california-homelessness-prediction-challenge
• Kaggle Notebook: https://www.kaggle.com/code/darcy59/cah-kim-neuralnet-kfold-with-tuning
• GitHub Repo: https://github.com/Darcy59/kaggle-california-homelessness
  Highlights
• Neural network approach with K-Fold validation
• Hyperparameter tuning + iteration tracking

Languages: Python, SQL
ML: scikit-learn, cross-validation, hyperparameter tuning, feature engineering
Analytics/BI: Tableau, visualization & storytelling

• Build stronger baselines and compare models (linear vs tree-based vs neural nets)
• Add error analysis and interpretability (feature importance / sensitivity checks)
• Include 1–2 visuals per repo (CV results, leaderboard, or key plots) for quick scanning