data-science-ml

Compare basketball performance fairly. Possession normalization, shrinkage, uncertainty, player comparisons.

Prioritize reviewable churn retention work. Calibrated models, eligibility rules, budget, audited task queue.

Explore salary estimates and uncertainty. Validated data, baseline, train/test split, prediction interval, feature controls.

Understand spending with editable categories. Rule categories, merchant overrides, monthly trends, standardized clustering.

[Data Science & Machine Learning] Prioritize reviewable churn retention work. Calibrated churn scores and eligibility rules; Capacity-limited review queue with cooldowns.

[Data Science & Machine Learning] Compare basketball performance fairly. Per-36-minute rates and scoring efficiency; Gamma-Poisson shrinkage and uncertainty intervals.

Adapt a small model for support routing. Fine-tuning, grouped splits, baseline, saved model, inference.

[Data Science & Machine Learning] Explore salary estimates and uncertainty. Synthetic salary regression with a median baseline; Separate training, calibration and test splits.

[Data Science & Machine Learning] Adapt a small model for support routing. Actual FLAN-T5-small support-ticket fine-tuning; Template-family splits and checkpoint selection.

[Data Science & Machine Learning] Understand spending with editable categories. Editable merchant rules and exact spending totals; Monthly summaries and normalized customer features.