I design platform libraries, runtimes and production systems across hardware and software boundaries. My work combines architectural direction with hands-on engineering in C++, Linux, IPC, memory and workload performance.
Experience
Nuro · Staff Software Engineer
January 2022 – present · Mountain View
Led the systems integration path for Nuro’s first autonomous miles on Lucid Gravity and dual NVIDIA DRIVE Thor: platform libraries, IPC, onboard/HIL orchestration, performance simulation and CPU/latency optimization.
Designed CPU–GPU ring-buffer abstractions spanning IPC and in-process communication, x86 and ARM, unified and discrete memory; used by Perception, Behavior, Remote Operations and logging. Built inter-Thor PCIe transport using kernel IOCTLs and DMA shared-memory synchronization.
Improved full-workload latency through real-time scheduling and core spillover, recovering more than 100 ms. Designed eBPF long-sleep tracing to reconstruct waker–sleeper dependencies across application and kernel boundaries.
Set logging and offloading architecture across onboard, depot and cloud. Simplified onboard design around append-only mission files, pre-I/O drop/hold decisions and memory/disk buffering; defined incremental delivery milestones.
Set accelerator-first architectural direction for Remote Operations video; enabled another engineer to own implementation trade-offs and the production transition.
Drive organization-level architecture/design reviews and code-review practices. Created Bugbot, an AI-native review service deployed in production across hundreds of PRs daily, with focused review guidance, evidence-aware findings and deterministic publishing.
Microsoft · Software Engineer, Azure Core OS
2020 – 2021 · Redmond
Built Windows Server host-update orchestration and kernel optimizations for VM downtime and availability across millions of Azure hosts.
Microsoft · Data Scientist Intern, Azure Compute
Summer 2019
Optimized Kusto hot-cache and cluster autoscaling policies, saving approximately $660K per month in internal analytics hosting costs.
Earlier: Microsoft Research Asia, Software Engineer Intern (2017); Morgan Stanley, Summer Analyst (2016).
Education
University of Washington · MS, Technology Innovation, Global Innovation Exchange · 2019 Xi’an Jiaotong University · BEng, Computer Science & Technology · 2018
Selected technical depth
C++ · Linux / eBPF · IPC / shared memory · CUDA memory interoperability · PCIe · io_uring · Real-time scheduling · Distributed simulation · AI-native developer tooling