CV

PhD

ETH Zurich

Computer Science

• Research on efficient evaluation of large language models (LLMs).
• First-author ICML 2026 paper in the first year.
• Two further projects to be submitted in early 2027.

MSc

ETH Zurich

Statistics

• Overall grade 5.85/6.00, with distinction; over 150 ECTS completed (90 required).
• Coursework: Guarantees for Machine Learning, Probabilistic Artificial Intelligence, Deep Learning in Scientific Computing, Probability Theory, Fundamentals of Mathematical Statistics, Functional Analysis I–II, Numerical Solution of Stochastic Ordinary Differential Equations.

BSc

London School of Economics (LSE)

Mathematics, Statistics and Business

• First Class Honours.
• Coursework: Stochastic Processes, Algorithms and Programming.

Cutting LLM Evaluation Costs with SySRs: A Bandit Algorithm that Provably Exploits Model Similarity

International Conference on Machine Learning (ICML 2026)

Successive Rejects with paired comparisons: hyperparameter-free, guarantees improve with model similarity; lower average error rate than all baselines on 15 benchmarks, and lower worst-case budget to identify the best model.

Checkable NTK Positivity and Finite-Width Gradient Descent for Scalar- and Vector-Valued PINNs with Strong-Form, Weak-Form, and Nonlocal Linear Constraints

Under review at the International Conference on Learning Representations (ICLR 2027)

Research Assistant

ETH Zurich

• Developed SySRs, a hyperparameter-free bandit algorithm that finds the best LLM with fewer evaluations by exploiting similarity between models (ICML 2026).

Research on physics-informed neural networks (MSc)

ETH Zurich

• Combining computer vision with physics-informed neural networks to estimate the underlying fluid dynamics (manuscript in preparation).
• Checkable positive-definiteness criteria for the limiting neural tangent kernel of PINNs with linear constraints, and high-probability guarantees for finite-width gradient descent (arXiv:2610.04357).

Teaching Assistant, Probability Theory (401-3601-00L)

ETH Zurich

• Weekly exercise sessions and grading for Prof. Igor Kortchemski’s graduate course.

Head of Diversity and Inclusion

LSESU Business & Investment Group

• LSE’s largest student society: organised diversity and inclusion events on campus and liaised with employers alongside its Women’s Advancement Division.

Peer Supporter

London School of Economics (LSE)

• Confidential peer support for students facing academic stress, peer pressure and interpersonal difficulties; trained in active listening, crisis intervention and boundary-setting under professional supervision.