I'm a senior applied scientist at Microsoft, leading large-scale ML systems in auction and autobidding at Bing Ads. My work sits at the intersection of causal inference, reinforcement learning, and foundation models — figuring out how to safely improve systems that can only observe the consequences of their own past decisions.

On the side, I teach graduate ML courses at Northeastern's Khoury College and have a decade-long track record of publishing in top-tier ML conferences — NeurIPS, AAAI, ACM RecSys.

Research

16 publications total · ~120 citations · Google Scholar · 2 granted US patents

Combining Open-box Simulation and Importance Sampling for Tuning Large-Scale Recommenders
CONSEQUENCES @ ACM RecSys 2024 oral · arxiv
Adaptive Mixture Importance Sampling for Automated Ads Auction Tuning
CONSEQUENCES @ ACM RecSys 2024 · arxiv
Leveraging Structured Biological Knowledge for Counterfactual Inference: A Case Study of Viral Pathogenesis
IEEE Transactions on Big Data, 7(1), 25–37, 2021
Integrating Markov Processes with Structural Causal Modeling Enables Counterfactual Inference in Complex Systems
NeurIPS 2019 · proceedings
Regularizing Fully Convolutional Networks for Time Series Classification by Decorrelating Filters
AAAI 2019, 33(01)
Crop Planning Using Stochastic Visual Optimization
IEEE VDS 2017, 47–51
Visual Bayesian Fusion to Navigate a Data Lake
IEEE FUSION 2016, 987–994
Multi-Sensor Visual Analytics Supported by Machine-Learning Models
IEEE ICDMW 2015, 668–674

Talks

SigIS: Simulator-Guided Importance Sampling for Recommender Tuning First-author oral · CONSEQUENCES @ ACM RecSys '24
Bari, Italy
Causal Transfer Random Forest in Production Microsoft Research seminar series
Redmond
Integrating Markov Processes with Structural Causal Modeling Poster · NeurIPS '19
Vancouver

Teaching

Spring 2023
DS 4400 — Machine Learning and Data Mining 1 (undergraduate) Visiting Lecturer · Khoury College, Northeastern University
Spring 2022
DS 5220 — Supervised Machine Learning and Learning Theory (graduate) Visiting Lecturer · Khoury College, Northeastern University
Fall 2021
DS 5220 — Supervised Machine Learning and Learning Theory (graduate) Visiting Lecturer · Khoury College, Northeastern University

Open Source

2025 – present
Python

crux — Agentic Tool Manager for Claude Code

Python CLI that orchestrates agentic workflows on top of Claude Code — sandboxed tool execution, OS-keychain secrets management, and MCP (Model Context Protocol) server + skills integration. Author and sole maintainer.

github.com/crux-cli/crux ↗

Writing

Hello, world

First post on the new site. Short note on what I plan to write about — causal inference, reinforcement learning, foundation models, and the occasional dispatch from teaching.