I'm Aishwarya, a Ph.D. Candidate and Computer Science Researcher at Iowa State University. My current research explores algorithmic and systems-level optimizations for neural networks, aiming to reduce computational and communication overhead while maintaining model accuracy. I am particularly interested in efficient data movement, parallel computing, memory management, and distributed optimization tailored for modern HPC platforms.
- working on accelerating neural networks on HPC systems.
- learning about Mixture of Experts (MoE), LLM Agents, and In-Context Learning.
- Graph Neural Networks (GNNs)
- High-Performance Computing (HPC) & Distributed Systems
- Deep Learning Optimizations & MoE Models
- Parallel Computing & Cloud AI
- Reading books from various genres and trying to explore as many topics as I can. Check out some of my favorite books here.
- Playing various racquet sports.
- Watching documentaries, especially on history and science.
❝ You're optimizing HPC like a pro, but somehow your debugging process still involves sacrificing print statements to the gods of segmentation faults.
You dream of reducing communication overhead—maybe start by not arguing with your own scripts at 3 AM. 🚀
Your GNNs are distributed, just like your sanity when debugging. You claim to love parallelism, yet your tasks are just a thousand threads of chaos with no mutex in sight.
And let’s face it, your code crashes more often than your motivation. At this point, your PhD isn’t about HPC—it’s about High Pain Computing. 💀
Is your HPC career dead yet 💀 or do you want me to vectorize the insults for even faster execution? 😈 ❞
— ChatGPT 4o