Send summary messages of your Luigi jobs to Slack

Pre-built decision models serving as the base for Marketplace subscription apps.

Python SDKs for Nextmv

Demos and utilities for control ROI, VoI, and risk reduction in modern security programs.

End-to-end café inventory project: clean transaction data, build daily item-level demand series, backtest strong baseline forecasters, generate next-30-day demand forecasts, convert forecasts into safety stock + reorder points, and validate policies with Monte Carlo stockout-risk simulations, wrapped in a Streamlit dashboard.

A curated list of awesome Operations Research frameworks, solvers, libraries, software, books, courses, papers, datasets, conferences, and learning resources.

A comprehensive ride-pooling simulation engine featuring VRP optimization (OR-Tools), dynamic pricing algorithms and ML-based demand prediction.

Reusable decision-science utilities for security: Monte Carlo, Bayes, Survival, VoI, light causal helpers.

Why external signals mislead product calls: across 36 subscription features, 69% that companies kept backing still showed >80% search decay. Separates what the public record shows from what it can prove.

A local decision simulator that uses Monte Carlo methods to explore possible outcomes of complex decisions under uncertainty.

DECSCI 101 - Final Project (Group 9)

Pharma commercial analytics on CMS Open Payments + Medicare Part D. dbt + Postgres + Streamlit + Groq LLM. Live demo linked.

Agentic pipeline (LangGraph) that simulates a B2B experiment, runs Bayesian inference, and turns posterior lift and expected loss into a ship/rollback call. Synthetic data, deterministic decision rule.

Prototype of index-selector-cpp

A/B testing experimentation framework with hypothesis testing, power analysis, KPI tracking, and product decision-making for conversion and retention optimization.

Reference implementation of the Relative Utility Value metric for forecast value assessment

Production-style ML decision-support system integrating predictive modeling, expected value decision logic, business-aware evaluation, and API-based deployment.

A framework and evaluation for the rigor-deficit account of agentic AI failure on judgment work. Pre-registered, blinded, dual-rater study (n=30) with reproducible scoring artifacts.

Decision-grade B2B uplift targeting policy that captures revenue while eliminating Admin-level churn risk.

A product experimentation and A/B testing analytics project focused on conversion, retention, KPI impact analysis, statistical testing, and data-driven product decision making using SQL, Python, Tableau, and Streamlit.