Summary
Andy Jost is a Principal Engineer specializing in CUDA Python and high-performance GPU computing, with 11+ years building production-grade simulation and acceleration systems for companies like NVIDIA, ASML, Siemens, and Synopsys. He combines deep systems and language expertise—Python, CUDA/C++, LLVM/MLIR, JITs, and AVX-512 assembly—with a strong background in scientific stacks (NumPy, cuPy, JAX) to deliver highly optimized, memory- and compute-efficient solutions. At Synopsys he architected a photolithography simulation used across a $100M+ product line, including custom JIT backends for CPU and GPU that dramatically improved throughput. Known for blending functional, logic, and distributed paradigms, he focuses on developer experience as much as raw performance, now bringing that ethos to CUDA tooling in the Python ecosystem at NVIDIA. He holds a PhD in Computer Science and earlier technical roots in lithography and materials science, which inform his practical approach to modeling, optimization, and domain-specific compilers.
11 years of coding experience
22 years of employment as a software developer
M.S./B.S., Biochemistry, Materials Science, GPA - 4.0, M.S./B.S., Biochemistry, Materials Science, GPA - 4.0 at University of Oregon
Doctor of Philosophy - PhD, Computer Science, 4.0, Doctor of Philosophy - PhD, Computer Science, 4.0 at Portland State University