sushi
Python
PyTorch
TensorFlow
NumPy
Pandas
Jupyter
Scikit-learn
Hugging Face
Docker
Kubernetes
AWS
PostgreSQL
FastAPI
Git
Linux
LangChain
OpenCV
VS CodeNeural Dreamscapes
Refik Anadol
Exploring the latent spaces of machine learning models through immersive data sculptures and live audio/visual performances.
About Me
Welcome to my corner of the internet. This is a living archive of my explorations, experiments, and everything in between. I'm passionate about artificial intelligence, machine learning, and building things that push the boundaries of what's possible.
Here you'll find my thoughts on AI research, technical deep dives, project write-ups, and insights from my journey in the world of technology.
An AIML Engineer currently at HTC Global Services in Bengaluru, with a career rooted in building intelligent systems from the ground up. The journey began with a B.Tech in Computer Science from United Institute of Technology, followed by three years as a Machine Learning Engineer at SwiftAnt, where the focus was on Retrieval-Augmented Generation for document intelligence, engineering high-availability PostgreSQL services, optimizing SQL procedures for enterprise data reporting, and managing CI/CD pipelines for production deployments.
In parallel, a Master's in AI & Machine Learning from BITS Pilani deepened the theoretical foundations across deep learning, natural language processing, and computer vision. A Google TensorFlow Developer Certificate further solidified hands-on expertise in building and deploying production-grade ML models.
The work sits at the intersection of research and engineering — taking ideas from papers and turning them into reliable, scalable systems. Core competencies span the full ML lifecycle: from data pipelines and model training with PyTorch and TensorFlow, to containerized deployments on Kubernetes and AWS.
Beyond the day job, this site serves as a living archive of explorations — blog posts, tutorials built from Jupyter notebooks, and project write-ups that document the process of learning in public.
Visual Inspiration
Refik Anadol
Exploring the latent spaces of machine learning models through immersive data sculptures and live audio/visual performances.
Manolo Gamboa Naon
Generative compositions that explore the mathematical beauty of fractal geometry and harmonic patterns in visual form.
Casey Reas
Software-based artworks that reveal the emergent beauty of simple rules interacting in complex systems and natural phenomena.
What I'm Into
Understanding what neural networks learn and how they make decisions. Exploring visualization techniques for model internals and attention patterns.
Using machine learning as a creative tool for generating visual art, music, and interactive experiences that blur the line between human and machine creativity.
Crafting beautiful, interactive visualizations that make complex data accessible and tell compelling stories through visual design.
How I Think
We can only see a short distance ahead, but we can see plenty there that needs to be done.
— Alan Turing
I believe in learning by doing. Every project, no matter how small, is an opportunity to deepen understanding and discover something unexpected. The most valuable insights often come from the experiments that don't work the way you planned.
My approach to learning combines rigorous mathematical foundations with hands-on experimentation. I think the best engineers are those who can move fluidly between theory and practice, understanding both the "why" and the "how."
Implement concepts from scratch before using libraries. Understanding the fundamentals makes you a better engineer and helps you debug complex systems.
Writing and teaching are the best ways to solidify understanding. If you can't explain something simply, you don't understand it well enough.
The most interesting problems are the ones where you don't know the answer going in. Stay curious, take risks, and don't be afraid to explore unfamiliar territory.