I am on the job market for full-time positions and open to opportunities in the industry. I would be grateful for any opportunities you might consider sharing via email (xushenbo857 {@ | at} gmail.com).

I completed my PhD at MIT. My research centers on the next generation of foundational models for data-driven dynamic decision-making for real-world problems. Specifically, I focus on devising methods that explain the consequences of actions:

• Reasoning on knowledge graphs, involving pre-training, reinforcement learning, agentic systems, and evaluation
• Causal inference, multivariate stochastic processes, uncertainty quantification, conformal inference, Bayesian active learning
• Applications in finance, crypto, medicine, sales, marketing, manufacturing, etc

Below is my industry experience

Machine Learning Research experience:
I worked at Scale AI, Inc and MIT-IBM Watson AI Lab as a researcher.

Quantitative Research experience:
I worked at Point72 Asset Management as a Quantitative Researcher and Liberty Mutual Investments and the MIT Quest for Intelligence) as a Research Assistant.

Selected Publications

• Do Larger Language Models Generalize Better? A Scaling Law for Implicit Reasoning at Pretraining Time by X. Wang, S. Tan, S. Xu, M. Jin, W. Wang, R. Panda, Y. Shen. Under review at the International Conference on Learning Representations 2026. arXiv preprint arXiv:2305.02373.

• Estimating heterogeneous treatment effects on survival outcomes using counterfactual censoring unbiased transformations by S. Xu, R. Cobzaru, S. Finkelstein, R. Welsch, K. Ng, and Z. Shahn. Revision under review at the Journal of Machine Learning Research. arXiv preprint arXiv:2401.11263. PR under review at EconML (part of PyWhy).

• Double/debiased machine learning for time-to-event outcomes under poor overlap by S. Xu, S. Finkelstein, R. Welsch, K. Ng, I. Tzoulaki, and Z. Shahn. Under review at the Neural Information Processing Systems (NeurIPS) 2026. arXiv preprint arXiv:2305.02373. PR under review at DoubleML.

• Foundational model-aided automated high-throughput drug screening using self-controlled cohort study by S. Xu, R. Cobzaru, S. Finkelstein, R. Welsch, and K. Ng. AI for New Drug Modalities (AIDrugX) at 38th NeurIPS, Vancouver, Canada, Dec 10-15, 2024. PR under review at ehrapy and EHRData.

• Can metformin prevent cancer relative to sulfonylureas? A target trial emulation accounting for competing risks and poor overlap via double/debiased machine learning estimators by S. Xu, B. Zheng, B. Su, S. Finkelstein, R. Welsch, K. Ng, and Z. Shahn. American Journal of Epidemiology, 2025, 194(2), 512-523. PR under review at DoubleML.

• Systematically exploring repurposing effects of anti-hypertensives by Z. Shahn, P. Spear, H. Lu, S. Jiang, S. Zhang, N. Deshmukh, S. Xu, K. Ng, R. Welsch, and S. Finkelstein. Pharmacoepidemiology and Drug Safety, 2022, 31(9), 944-952

• Contributor to QuantitativeFinance-Bench: A State-Aware Interactive Benchmark for Financial Agent Tasks, 2026.