Senior Software Engineer at OpenTeams
Currently contributing to the PyTorch ecosystem, with a focus on high-performance GPU kernels, compiler infrastructure, and scientific computing integration.
I bridge the worlds of computational quantum chemistry and high-performance software engineering, developing tools that make large-scale scientific computation faster, more scalable, and more maintainable.
| Project | Description | Tech |
|---|---|---|
| PyTorch | Core deep learning framework — GPU kernels, Inductor compiler, and performance optimization | C++, Python, CUDA |
| Einsums | C++20 tensor library for single-node and GPU computations | C++20, CUDA / HIP |
| Psi4 | Open source quantum chemistry | C++20, Python |
- GPU kernel development and tensor compiler optimization (PyTorch Inductor)
- Tensor rank reduction (DLPNO, THC)
- Task-based parallelism & custom AGAS designs
- Performance benchmarking & profiling tools
- Scientific Python/C++ interoperability
"Efficient science needs efficient code."