Aarush Agarwal
ML Engineer Intern @Shopify | AI @CMU
Los Altos, California • Open to opportunities
About
Machine Learning Engineer and AI researcher with hands-on experience in high-performance computing optimization and practical AI applications. Currently pursuing Artificial Intelligence at Carnegie Mellon University while working as an ML Engineer Intern at Shopify.
Passionate about bridging the gap between theoretical AI concepts and real-world implementations, with a focus on CUDA optimization and graph neural networks for particle physics applications. My research involves optimizing machine learning algorithms for high-energy physics, achieving up to 40x speed improvements over industry-standard libraries.
From co-founding educational technology startups to conducting cutting-edge research in particle physics, I'm always excited to tackle complex problems that make technology more accessible and impactful. I blend entrepreneurial experience with deep technical expertise to create meaningful solutions.
Experience & Education
Machine Learning Engineer Intern
May 2025 - PresentShopify
Developing a more robust machine learning model to detect buyer fraud on Shopify. Used to prevent fraudsters from making transactions and improve platform security.
ML & CUDA Researcher
August 2024 - PresentCarnegie Mellon University Cosmology Lab
Conducting research under Professor Mateo Cremonesi. Co-authoring research on optimizing the K-Nearest Neighbors algorithm using advanced spatial partitioning, achieving up to 40x speed improvements over FAISS in low-dimensional spaces. Accelerated KNN runtime by 10% by migrating Python autograd operations to high-performance C++ and CUDA extensions integrated with PyTorch JIT.
Key Achievements
High-Performance ML Research
Achieved up to 40x speed improvements over FAISS in low-dimensional vector spaces through novel K-Nearest Neighbors algorithm optimization using advanced spatial partitioning techniques.
Educational Technology Impact
Co-founded MV Test Tracker, revolutionizing academic scheduling for Mountain View High School and improving student mental health through intelligent test scheduling optimization.
Competitive Programming Excellence
Led two teams to semifinalist positions in the Mathematics and Technology for Computer Science (MtFC) competition in 2022 while serving as Vice President of AI Club.
CUDA Optimization Research
Developing cutting-edge CUDA extensions for particle physics applications, optimizing graph neural networks for particle collision detection in CERN experimental data.