Software Engineer - C++ / GPU Virtualization / Cloud Infrastructure
USA, UTILITIES SERVICES ALLIANCE, INC.
Brisbane, United States of America
yesterday
Role details
Contract type
Permanent contract Employment type
Full-time (> 32 hours) Working hours
Regular working hours Languages
EnglishJob location
Brisbane, United States of America
Tech stack
Systems Engineering
C++
Cloud Computing
Nvidia CUDA
Software Debugging
Distributed Systems
Memory Management
General-Purpose Computing on Graphics Processing Units
InfiniBand
Machine Learning
PCI Express
Performance Tuning
System Programming
Virtualization Technology
Large Language Models
Machine Learning Operations
Job description
- Own the performance optimization of the core C++ virtualization library, squeezing every microsecond out of the stack.
- Drive R&D into complex problems like GPU oversubscription, checkpointing, and distributed cluster orchestration.
- Implement support for next-generation GPU architectures, requiring a deep understanding of the entire system from driver to user-space.
- Tackle high-stakes systems-level debugging and root-cause analysis in live production environments.
- Identify and resolve performance bottlenecks specifically within large-scale Machine Learning and LLM training workloads.
Requirements
- Exceptional skills in modern C++ (Systems level) with a focus on memory management, concurrency, and low-latency execution.
- A strong grasp of GPU computing (CUDA/ROCm), drivers, or virtualization technologies.
- Comfortable working close to the metal, including knowledge of PCIe, NVLink, or InfiniBand.
- A background working in Trading Companies/Hedge Funds/HFT is highly advantageous (either in industry or academic internships.
Keywords: Systems Engineer / C++ / GPU Virtualization / CUDA / Performance Optimization / Low-Level Programming / Distributed Systems / Cloud Infrastructure / Kernel / Drivers / ML Infrastructure / High-Performance Computing (HPC)