Summary
Yifei Xu is a software engineer with six years of experience building AI infrastructure and high-performance ML runtimes, now working on PyTorch at Meta Superintelligence Labs after contributing to Core ML Runtime at Google. He brings deep systems and numerical expertise from a Ph.D. in reservoir simulation, where he developed advanced fracture modeling and grid-generation algorithms, and has translated that into production work on distributed training and serving for TensorFlow, JAX, and IFRT. His background spans C++, Python, JAX, TPU work, and cloud-native tooling—including Terraform, Kubernetes, and microservices—giving him fluency across research-grade algorithms and production ML platforms. At ExxonMobil he led small engineering teams to deliver simulation and optimization software, demonstrating both technical leadership and domain-driven product delivery. Comfortable moving between low-level performance engineering and higher-level infrastructure, he’s equally at home optimizing distributed runtimes or designing resilient deployment stacks. Based in the Bay Area, he blends rigorous academic modeling experience with hands-on cloud and ML systems engineering.
6 years of coding experience
16 years of employment as a software developer
Doctor of Philosophy (Ph.D.) Petroleum Engineering Reservoir Simulation, Doctor of Philosophy (Ph.D.) Petroleum Engineering Reservoir Simulation at The University of Texas at Austin
Bachelor of Science (B.S.) Thermal Engineering, Bachelor of Science (B.S.) Thermal Engineering at Tsinghua University
High School Science and Engineering, High School Science and Engineering at Attached Middle School of Henan Normal University
English, Chinese