7 hours ago
Base Salary
$244k - $413k/yr
Responsibilities
- Own step time, scaling efficiency, and long-run stability for foundation-model training across thousands of accelerators.
- Partner with modeling, infrastructure, simulation, data-mining, and robotics teams to resolve complex performance bottlenecks.
- Turn performance improvements into modular components and defaults reusable across models and workloads.
- Extend performance tooling and metric coverage while maintaining low instrumentation overhead for production workloads.
- Build agents that reproduce jobs, verify numerical parity, and identify genuine performance regressions.
Requirements
- Master's or PhD in computer science, computer engineering, or electrical engineering, or equivalent practical experience, with 8+ years of industry experience, primarily in performance engineering.
- Hands-on production experience optimizing GPU training or inference workloads at multi-node scale.
- Demonstrated ability to diagnose and fix performance problems in production systems, including systems owned by other teams, with gains measured in production.
- In-depth understanding of computer architecture and high-performance computing.
- Strong C++ and Python skills and command of data structures and algorithms.
- Preferred experience includes tuning CUDA or Triton kernels, using PyTorch, training across thousands of accelerators, handling multi-node failure modes, and building profiling, tracing, or benchmarking infrastructure.
- Preferred qualifications include experience with CUPTI or Perfetto, LLM-agent workflow automation, autonomous driving or robotics, and publications or open-source contributions in ML systems or performance.
Benefits
- Competitive compensation package with bonus, equity, and benefits.
- Supportive and engaging work environment with computational resources and infrastructure.
- Opportunity to work on cutting-edge autonomous driving and robotics technologies.
- Snacks, lunches, dinners, and fun activities.
Categories
About XPENG
XPENG designs, manufactures, and sells smart electric vehicles for consumers, integrating in-house advanced driver-assistance systems and connected in-car software. Founded in 2014 and headquartered in Guangzhou, it operates plants in Zhaoqing and Guangzhou, maintains a European HQ in Amsterdam, and develops eVTOL aircraft via XPENG AEROHT.