Vik Shirvaikar
AI/ML Engineer
PhD, Statistics and Machine Learning, University of Oxford
Bachelor's, Math and Economics, University of Texas at Austin
Confusion Matrix has moved to beehiiv!
Read and subscribe at the link above! Old archived posts can be found below.
Industry R&D
I am currently the Founding AI/ML Engineer at Gyre Energy, an early-stage startup developing automated control systems to make cooling infrastructure more energy-efficient. My work focuses on designing and calibrating physics-informed digital twins to optimize operational strategies.
Selected Papers
My doctoral research focused on Bayesian prediction, causal inference, and model uncertainty, with applications to healthcare, public policy, and large language model (LLM) evaluation.
- As lead author
- Prediction-powered machine learning for model selection and uncertainty
- Targeting relative risk heterogeneity with causal forests
- Rethinking recidivism through a causal lens
- As contributing author
- A decision-theoretic framework for uncertainty quantification in epidemiology
- Do Bayesian neural networks actually behave like Bayesian models?
- A critical review of causal reasoning benchmarks for large language models
See below, or download the file.