Home Experience


  • Since August 2022

    Doctoral Candidate

    MPI for Maths in the Sciences.
    Max Planck School of Cognition.

    Jürgen Jost, Peter Dayan

    Started my PhD at the MPI for Mathematics in the Sciences in Leipzig under Jürgen Jost and Peter Dayan. We are planning to look at the underlying neural dynamics of the rodent brain, and investigate them from an information theoretic and dynamical systems perspective. We want to check if the coherence across multiple brain regions is a byproduct of the communication itself, or is it the other way around,as it has been classically understood.

  • May-July 2022

    Lab Rotation

    MPI for Biological Cybernetics.

    Peter Dayan

    I worked with Peter and Oleg at the MPI for Biological Cybernetics in Tuebingen. We investigated why path dependence might arise in actor critic reinforcement learning models. We used transformers to gain insights into the behavior of classically trained RL agents, and also used hyperencoders to check if the latent space of differing trajectories could be identified. Check out this blog post for more information.

  • Jan-March 2022

    Lab Rotation

    MPI CBS.

    Christian Doeller

    I worked with Christian and Stephanie at the MPI for Human, Cognitive and Brain Sciences in Leipzig. We investigated if we could pinpoint if semantically similar and dissimilar information encoded in the brain can be separated by their relative embedding distances in the hippocampus. To that end, we used word embeddings from state of the art machine learning models to code for novel concepts and performed RSA analyses in the hippocampus to check if they were being represented as we thought. Sadly, we did not find anything significant.

  • Sep-Nov 2021

    Lab Rotation

    MPI for Mathematics in the Sciences.

    Jürgen Jost

    I worked with Jürgen and Michelle at the MPI for Mathematics in the Sciences in Leipzig. We investigated if we could use the ricci-forman curvature of hypergraphs and a mutual information based metric of connectedness to reduce high dimensional fMRI data to manageable proportions. We then used the curvature for the nodes as a proxy for summarizing the entire data and correlated it to iq scores of participants. We found that the prediction worked better for graphs rather than hypergraphs, but Michelle is working on this project to get to the bottom of it.

  • Jan 2020-July 2021

    Research Assistant

    CIMeC, University of Trento.

    Roberto Bottini

    I was employed as a research assistant at CIMeC in the University of Trento. I worked on a bunch of projects, but mostly on the similarities and dissimilarities of color and emotion representation in the brain, hypothesized by us to involve cognitive maps with Jean Victor. You can check out this blog post for more information. I also helped Guiliano to code up and collect data for a MEG experiment that he is doing for his PhD. The experiment involved trying to see if we could identify hexagonal grid coding in the hippocampus using MEG.

  • Aug 2018-Nov 2019

    Member of Technical Staff


    Oracle Cloud Infrastructure

    I worked on the OCI team for Oracle migrating from the old micro-services based model to a new cloud based model. I was mostly involved with increasing prediction accuracy for the infrastructure monitoring software, and also worked on a technology to mark inefficient queries on the fly using machine learning.

  • Aug 2016-July 2018


    Indian Institute of Science, Bangalore

    Department of Computer Science and Automation

    I was ranked 7th in the all India wide held GATE examination, for which I got a seat at the prestigious Computer Science department at IISc Bangalore. I did my master's thesis under Prof. Sridharan Devarajan, in conjunction with the Center for Neuroscience. We worked to develop a multi-modal model to predict conversion from MCI to Alzheimer's. We also tried to see if it was possible to combine data from multiple centers using transfer learning to increase the statistical power. You can take a look here if you want to read up on it.

  • Aug 2012-Apr 2016


    Indian Institute of Engineering, Science and Technology, Shibpur

    Department of Computer Science and Technology

    I did by Bachelor's in computer science at IIEST Shibpur. I did my bachelor's thesis under Prof. Jaya Sil.We investigated deep learning and n-gram based approaches to perform document classification. You can take a look here if you want to read up on it.

Biting comic about science from nathiel Biting comic about science from nathiel Biting comic about science from nathiel
--> -->