cv

PhD

University of Amsterdam

Physics

MSc

University of Amsterdam

Physics and Astronomy (Theoretical Physics), cum laude

B.Tech

Indian Institute of Technology Madras

Engineering Physics

Surveillance Analyst

Goldman Sachs, Bangalore, India

Developed ML models for anomaly detection in financial data. Manager: Prof. Howard Karloff

Summer Intern

Goldman Sachs, Bangalore, India

Implemented fast Personalized PageRank in a MapReduce framework. Mentor: Dr. Koushik Balasubramanian

TopoFisher: Learning Topological Summary Statistics by Maximizing Fisher Information

Under review

We learn topological summary statistics for cosmological inference by maximizing the Fisher information of persistent homology features.

A Transformer Architecture Alteration to Incentivise Externalised Reasoning

Preprint

We modify the transformer architecture to encourage models to externalise their reasoning in the chain of thought.

The Intrinsic Dimension of Prompts in Internal Representations of Large Language Models

Transactions on Machine Learning Research (TMLR)

We investigate the relationship between the geometry of token embeddings and their role in next token prediction.

Probing Geometry of Next Token Prediction Using Cumulant Expansion of the Softmax Entropy

ICML 2025: Workshop on High-dimensional Learning Dynamics

We introduce a cumulant-based framework to analyze how transformer models capture higher-order statistics during next-token prediction.

Persistent Topological Features in Large Language Models

ICML 2025 (Poster)

We use zigzag persistence to describe data undergoing dynamic transformations across layers of LLMs.

Cosmology with Persistent Homology: a Fisher Forecast

Journal of Cosmology and Astroparticle Physics

We apply persistent homology to dark matter halo catalogs and find tighter constraints on cosmological parameters.

Verified Mechanisms

Independent research project, London

Research lead. LLM agents propose and formally verify theorems about internal computations in transformers. Secured project funding in the grantmaking.ai Launch Round.

ML Alignment & Theory Scholars Program (MATS 9.0)

Simplex AI Safety, Berkeley & London

Do Alignment Pretraining if you want your Linear Probes to Work Better. Mentors: Adam Shai & Paul Riechers

Mentorship for Alignment Research Students (MARS 3.0)

Geodesic Research, Cambridge

Architectures for Increased Externalisation of Reasoning. Mentors: Puria Radmard, Cameron Tice & Edward Young

Workshop: Interpretability in LLMs using Geometric and Statistical methods

University of Amsterdam

Organized with Jan Pieter van der Schaar

Sander Bais Prize for Academic Merit

Institute for Theoretical Physics Amsterdam

For exceptional academic performance in the master's program

ACM ICPC World Finals (Honorable Mention)

ACM

Represented India in the international collegiate programming contest

International Olympiad in Informatics Training Camp

IARCS

Selected after ranking in top 25 in the Indian National Olympiad in Informatics

Exploring the Spectral Theory/Topological Strings Duality (Master's Thesis)

This thesis analyzes the ST/TS duality, linking topological strings on toric Calabi-Yau manifolds to quantum mechanical spectral theory.

Real Space Renormalization and Applications to Machine Learning (Bachelor's Thesis)

We explore the application of renormalization group (RG) theory to understand the success of Restricted Boltzmann Machines (RBMs).