Bayesian Statistics | Data Assimilation | Predictive Modeling
Ph.D. Candidate in Computational Science at the University of Oslo, specializing in Bayesian data assimilation and statistical inference — applied to Arctic glacier mass balance modeling and prediction under uncertainty.
I'm a Marie-Curie funded researcher working on the CompSci Project at the Department of Geosciences, University of Oslo. My work focuses on observing system simulation experiments (OSSE) through observational data assimilation in Arctic glacier modeling.
With a multicultural academic journey, I bring diverse perspectives to computational climate science. I have conducted extensive fieldwork across diverse glacial environments, including Svalbard, the Italian Alps, the Chilean Andes, and Iceland—bridging computational methods with hands-on field observations.
I develop transferable quantitative methods for prediction and decision-making under uncertainty. These skills apply across industries—from climate science to finance, energy, and autonomous systems.
Core expertise in combining observational data with predictive models using Bayesian inference. These methods enable optimal forecasting, uncertainty quantification, and real-time model updating—applicable to any domain with complex data streams.
Applying these methods to improve glacier mass balance predictions in Svalbard. Demonstrated ability to work with large-scale environmental data, satellite observations, and complex physical models in a challenging real-world domain.
Combining statistical methods with machine learning for parameter estimation, model calibration, and pattern recognition. Bridging traditional Bayesian approaches with modern ML techniques.
Developing ensemble-based data assimilation methods for improving glacier mass balance predictions using Bayesian inference and uncertainty quantification.
View Project →Strategic integration of CCUS technologies within global Environmental, Social, and Governance frameworks for sustainable climate solutions.
View Project →An exploration of why LLM agents should write their own memory rather than read human-built knowledge bases — covering MemGPT, Reflexion, A-MEM, and the harness engineering paradigm.
View Project →A reusable Claude Code skill for evaluating agentic AI systems: rubric design, bias-aware judge prompts, per-step attribution, and content-addressed caching for incremental re-evaluation.
View Project →Cao W, Aalstad K, Schmidt LS, Westermann S, Schuler TV · Journal of Glaciology, 2025 · DOI
Cao W, Schmidt LS, Aalstad K, Westermann S, Schuler TV · EGU General Assembly, 2024
Cao W, Schmidt LS, Aalstad K, Westermann S, Schuler TV · IUGG General Assembly, 2023
Led cross-functional project teams in one of China's largest real estate developers, overseeing planning, coordination, and delivery of large-scale development projects. Developed strong skills in stakeholder communication, resource management, and decision-making under pressure.
Acted as a direct bridge between union members and the union, ensuring that individual concerns — from employment rights to workplace conditions — were heard, represented, and addressed through collective dialogue and negotiation within the Norwegian academic system.
Served as the elected representative for PhD candidates at the University of Oslo, facilitating open communication between doctoral researchers and university administration on matters of academic governance, working conditions, and researcher welfare.
Interested in collaboration on data assimilation, glaciology, or climate research?
Department of Geosciences, University of Oslo
Sem Sælands vei 1, 0371 Oslo, Norway