About Me
I am an interdisciplinary researcher with research interests at the intersection of Machine Learning, Computer Science, and Computational Neuroscience. I am actively seeking roles in Data Science, Machine Learning, and AI, both in industry and research.
I hold a B.Sc. (Hons) in Physics from Miranda House, University of Delhi, and an M.Sc. in Life Science Informatics from the University of Bonn, Germany, after which I worked as a Research Scientist in the Learning in Early Childhood research group, developing preprocessing pipelines for multimodal data, including EEG-fNIRS, OPM-MEG, and EEG-fMRI, which led to insights into human cognitive development as well as vision.
During my time at FAU Erlangen-Nuremberg, I focused on representational alignment between humans and machines using a computer graphics-based approach to understand the brain, including reconstructing faces with the Basel Face Model. My broader research interests also include applying deep learning to neuroimaging, and I am passionate about building bridges between neuroscience and AI.
| Contact: LinkedIn |
Activity
Here are some stats related to my recent activity on GitHub:
🌟 Highlights
✨ Fun Fact: I love watching action movies and programming in my free time.
Activities and Highlights
VMV 2025
I attended VMV 2025 as a helper and organizer at FAU Erlangen-Nürnberg.
UniReps 2025 Reviewer
I served as a reviewer for the UniReps 2025 conference.
M2L Summer School, Croatia
I attended the M2L Summer School in Croatia from September 8–12, 2025. It was a fantastic experience meeting scientists and researchers from across the field. 300 participants were selected from 1700 applicants, covering topics from transformers to diffusion models.
Cohere Summer School Online
I attended the Cohere Summer School Online and remain connected with the community!
ICLR 2025 Presentation
I had the opportunity to present my work at ICLR 2025, one of the premier conferences in machine learning, held in Singapore. During the poster session, I showcased my research titled “Computer Graphics from a Neuroscientist’s Perspective” (PDF), engaging with domain experts from both computer science and neuroscience. The constructive feedback and interdisciplinary discussions significantly informed and shaped the next phase of my project.
Read my full ICLR 2025 recap →
Society for Neuroscience (SfN) Chapter, Singapore
I attended the Society for Neuroscience chapter in Singapore, where I participated in neuroscience outreach, helped organize local events, and fostered interdisciplinary collaboration between computational and biological sciences.
🛠️ Technologies & Tools
Languages
Frameworks & Libraries
Tools
💻 IDEs/Editors
🖥️ Libraries
Things I find fascinating
- The Twin Paradox - a thought experiment in special relativity where one twin travels near light-speed and returns younger than the twin who stayed on Earth, illustrating that time dilation is real and asymmetric between the two reference frames. It’s not just theoretical: after 340 days on the ISS, astronaut Scott Kelly returned about 5 milliseconds younger than his identical twin, Mark Kelly.
- ALS (motor neuron disease) - a progressive neurodegenerative disease that destroys the motor neurons controlling voluntary muscle movement; it remains incurable, though a small number of patients, most famously Stephen Hawking, have survived for decades rather than the typical 2-5 years.
- Gene expression control - not every cell with the same DNA does the same thing; regulatory mechanisms like transcription factors, epigenetic markers (e.g. DNA methylation, histone modification), and RNA splicing determine which genes actually get turned into proteins, when, and in which cell type.
- Molecular docking & simulation - computational methods that predict how a small molecule (e.g. a drug candidate) binds to a target protein, and simulate the resulting structure’s dynamics over time; central to modern structure-based drug design, building on the same energy-landscape ideas behind protein folding.
- Predictive coding & the free energy principle - Karl Friston’s theory that the brain isn’t a passive receiver of sensory input but a prediction machine, constantly generating a model of the world and updating it only when reality violates its expectations; perception becomes controlled hallucination, corrected by prediction error, an idea now quietly shaping how we think about learning in artificial neural networks too.
📝 Latest Blog Posts
- Visit to M2L Summer School
- My experience at ICLR 2025
- Remote Development using Pycharm
- A Necessity, A Comfort or Both
🤝 Let’s Connect
Some talks which I suggest to watch
- Encoding and Decoding speech from the brain
- ‘Godfather of AI’ predicts ALL jobs will be in ‘wiped out’ by AI
-
[Artificial Intelligence helps to detect diseases Daniel Rückert is Humboldt Professor for AI](https://www.youtube.com/watch?v=db62ot5S21c)
Courses I suggest to audit
- Introduction to Machine Learning
- Computational Visual Perception - I was also a part of this course as a Teaching Assistant
- Visual Computing for the Life Sciences - Offered by Prof. Dr. Thomas Schultz and is one of the best courses I ever attended. I must say that Prof. Schultz is one of the best course instructors I have met or learned from. His slides were always clear, structured and very easy to understand. They made hard concepts in Computer Science available to an audience of non-computer scientists at the Master level which was nothing short of great for the students.
- Introduction to Deep Learning
- Introduction to Neuroscience
- Introduction to Brain Computer Interfaces
Podcasts and Videos I suggest watching
- Consciousness, reasoning and the philosophy of AI
- Pawan Sinha with Project Prakash
- Peter Dayan, The Marriage of Natural & Engineered Reinforcement Learning - RLC 2025
- Previous work from Peter Dayan
- Machine Learning for the Sciences by Klaus Robert Muller
- The Brain that Changes itself
- How biocomputing works and Matters for AI
- The future of NeuroAI
- Quantum Computing
- WayMo - autonomous vehicles are possible :)
- Gary Marcus on AGI
- Ray Kurzweil, the wonderful and terrifying implications of computers that can learn
- Education in transition – we are not applying our brains to it?
- Inventing Liquid Neural Networks
- Prof. Bishop’s new Computer Vision Notebook
- The day after AGI
- The Brain that Changes Itself - the first documentary which inspired me about the plasticity of the brain and how are abilities are not fixed over life. We can change it over time with Neurofeedback, also use BCIs (brain computer interfaces) for Neurofeedback!
- Go to Zero
-
[Artificial Intelligence helps to detect diseases Daniel Rückert is Humboldt Professor for AI](https://www.youtube.com/watch?v=db62ot5S21c) - Love the Science
- The Hardest Problem AI Ever Solved, with Google DeepMind CEO - the best moment to use AI
- How AI Cracked the Protein Folding Code and Won a Nobel Prize
- Boltzmann Machines - one of my favorite topics in ML, and part of my semester research
- Lecture 11/16: Hopfield Nets and Boltzmann Machines - from Coursera; Hopfield networks are foundational to understanding energy-based models
- Biggest Breakthroughs in Biology and Neuroscience: 2025
Articles, which can also be listened to as a Podcast
Movies I suggest
a. A Beautiful Mind (about John Nash and Game Theory)
b. The Imitation Game
c. Theory of Everything
d. Inception
e. Shutter Island
f. Sully
Recommended Reading
A spelling device for the paralysed (Nature, 1999)
One of the foundational papers in Brain–Computer Interfaces (BCIs).
This work showed that completely paralyzed patients could communicate using EEG-based brain signals.
A landmark study demonstrating early non-invasive brain–computer communication for locked-in patients.
Habilitation Thesis — Simone Frintrop (2014)
A habilitation thesis from the University of Bonn (now Professor at the University of Hamburg), focused on computational visual attention systems. Frintrop’s work bridges biological and computational models of visual attention, covering saliency detection, top-down and bottom-up attention mechanisms, and their application to object detection and robot vision. Her earlier PhD work produced VOCUS — a widely cited visual attention system — and this habilitation extends those ideas further. A must-read for anyone interested in how machines can learn to “look” the way humans do.
Hierarchical Neural Networks for Image Interpretation — Sven Behnke (2003)
One of the greatest theses in the field of neural networks and computer vision. Behnke’s work laid foundational ideas for hierarchical convolutional architectures, influencing how we think about deep learning for visual processing long before the deep learning era took off.
A visionary work on hierarchical neural networks that anticipated many ideas central to modern deep learning.
Assembly of Protein Tertiary Structures from Fragments — Simons, Kooperberg, Huang & Baker (1997)
The founding paper of Rosetta, David Baker’s protein structure prediction method. What makes it interesting from a physics standpoint is the framing: folding is posed as a search for the global minimum of an energy landscape, tackled with Monte Carlo fragment assembly rather than full atomistic simulation — a statistical-mechanics approach to what is otherwise an intractable combinatorial problem. Baker later shared the 2024 Nobel Prize in Chemistry for this line of work, alongside the AlphaFold team.
Physics-first protein folding: energy landscapes and Monte Carlo sampling, decades before deep learning took over the field.
Books I suggest
a. The Singularity is Near by Ray Kurzweil
b. An Investigation of the Laws of Thought by George Boole on Boolean Algebra
c. Gödel, Escher, Bach: An Eternal Golden Braid by Douglas Hofstadter
d. Thinking, Fast and Slow by Daniel Kahneman
e. The Brain from Inside Out by György Buzsáki
f. How the Mind Works by Steven Pinker
Thank you for visiting my profile!