👨‍💻 Ongoing Work

• Building efficient algorithms for training Machine Learning models [1]

• Developing a cross-platform 128-bit floating-point dtype for NumPy

  → numpy-user-dtypes: quad dtype

🚀 Learning & Exploring

• LLVM & Compiler Development
• Kernel-level optimizations
• Distributed Systems & large-scale compute

🌱 What I’m focused on

Pushing the boundaries of numerical computing, performance engineering, and low-level systems, while building tools that make advanced computation accessible to everyone.