XGBoost & LightGBM

What it is

Gradient-boosted decision trees. XGBoost and LightGBM dominate Kaggle and most real-world tabular ML problems — fast, accurate, well-understood, easy to deploy.

How Vaaani uses it

• Lead scoring and conversion prediction
• Churn modeling for SaaS retention
• Fraud detection with class imbalance handling
• Feature importance analysis to inform business decisions

Why it makes the cut

If your data lives in tables and rows, gradient boosting beats deep learning 9 times out of 10. I reach for XGBoost first on every tabular brief.

Sample code

import xgboost as xgb

dtrain = xgb.DMatrix(X_train, label=y_train)
params = {"max_depth": 6, "eta": 0.1,
          "objective": "binary:logistic"}
model = xgb.train(params, dtrain, num_boost_round=200)