AMAN SWAR

AI Systems Engineer & Researcher building high-performance ML frameworks, self-supervised learning libraries, and next-gen AI infrastructure

pushing the limits of optimization, scalability, and intelligence.

View My Work CONTACT ME
Swipe Down
Aman Swar

AI Research Engineer

About Me

I’m an AI Systems Engineer & Researcher who’s downright obsessed with building stuff that makes machines think harder and run faster. I mess around with high-powered machine learning frameworks, cooking up various deep learning libraries and dreaming up next-gen AI infrastructure that can handle whatever I throw at them. I love getting my hands dirty with the low-level guts—think low-level optimizations, CUDA and GPU programming wizardry—and scaling AI so it doesn’t just crawl, it flies. Outside the tech cave, I’m a huge cheerleader for open-source AI. I’m always tossing ideas into the wild—blogs, projects, whatever—because sharing’s half the fun. My big dream? Making AI accessible to all.

2+
Years Coding
5+
Major Projects
2
Ongoing Research
2
Open Source

Technical Skills

Programming Languages

Python 80%
C++ 60%
C 40%
CUDA 40%

AI/ML frameworks

PyTorch 80%
TensorFlow 20%
Hugging Face Transformers 40%
SciKit-Learn 30%

Parallel Programming

CUDA Programming 60%
cuBLAS 20%
OpenMP 5%

Machine Learning

Deep Learning 70%
Natural Language Processing 60%
Computer Vision 70%
Representation Learning 80%

Computer Science Fundamentals

Algorithm 40%
Data Structures 60%
Operating Systems 60%
Computer Architecture and Organization 70%
GPU Architecture 70%
Parallel Programming 40%

Featured Projects

SEE IN DETAIL
Project 1

KernelLab

Making GPUs go Brrrrr.... High-Performance CUDA Kernels for Deep Learning & HPC

C++ CUDA
Project 2

RetinaSys

Deep Learning pipeline for detecting Diabetic Retinopathy

PyTorch Computer Vision
Project 3

SearchSphere

Windows Search, but it actually works

Python Transformers FAISS PyTorch

RetinaSys: A Real-Time, Interpretable Self-Supervised Framework for Generalized Diabetic Retinopathy Detection

September 2024 - present
  • Research under Dr.K.Vijayalakshmi
  • A novel system for the detection of diabetic retinopathy designed to address key gaps. RetinaSys is not only about a new model but also integrates explainable AI (xAI) for interpretable decision-making, generalizes across diverse demographics through robust training on varied datasets, and is optimized for real-time deployment in resource-constrained environments.
  • Hoping to publish paper in top-tier conferences