Building
intelligent
systems.
I work at the intersection of AI, research and computing, building systems across generative AI, LLMs, intelligent agents, computer vision and computational machine learning.
From ideas
to intelligent
systems.
Generative & Intelligent AI
Exploring how generative models can reason, use tools, learn from feedback, and operate as adaptive intelligent systems.
Scientific & Computational ML
Exploring computational paradigms for scientific machine learning, physics-informed neural networks, latent reasoning, and mathematical modeling.
AI Systems & Infrastructure
How can we make AI systems efficient to train, deploy, and operate across increasingly complex computational environments?
Algorithms & Computational Methods
How can mathematical formulations and computational algorithms make intelligent systems more robust, interpretable, and capable?
Questions I'm
following.
Areas I'm actively learning, experimenting with, or thinking about next.
Efficient LLM Training
What new computational paradigms can make intelligent systems more mathematically grounded, efficient, and capable?
Agent Architectures
How can agents use tools, skills, memory, self-learning loops, and graph engineering to operate with greater autonomy and less UI?
Reinforcement Learning
How can learning through feedback improve model behavior, reasoning, and decision-making?
Inference & Optimization
How can we make intelligent systems faster, more memory-efficient, and more efficient at inference?
Parallel & High-Performance Computing
How can parallel algorithms and high-performance computing accelerate increasingly complex intelligent systems?
I like working
between the two.
Research gives me the questions. Engineering gives me a way to test them.
I like working between research and production—turning ideas into working systems while using real-world constraints to uncover new questions worth researching.
A few things
I've shipped.
Publications
Explore ↗EngineeringProjects
Explore ↗Open SourcePackages
Explore ↗+ Systems
Research engineering
Building,
learning,
figuring things out.
I'm Shreyansh, a Computer Science student exploring intelligent systems, scientific machine learning, and computational methods. I like working between research and production-developing ideas through experiments, building systems around them, and letting real-world constraints expose new questions. I'm especially interested in understanding how intelligent systems can reason, learn, and become more capable through better algorithms and computation. When I'm away from the screen, you'll usually find me playing cricket, listening to music, or walking in nature. I like keeping things simple, staying curious, and making time to slow down.
More about me↗