Performance Engineer

The role

You will build optimization layers and profiling infrastructure to maximize the efficiency of large-scale training and inference workloads. This role serves as a bridge between frontier AI models and high-performance computing infrastructure.

What you will do

• Analyze model architectures to optimize mathematical operations on hardware
• Implement efficiency techniques like quantization, model pruning, and distillation
• Streamline data pipelines for high-performance accelerators
• Use observability tools to trace end-to-end execution of training jobs
• Build and maintain real-time dashboards for system metrics
• Monitor and justify the utilization of AI accelerators
• Develop scaling laws and internal tools to predict system behavior
• Work with AI Researchers to adapt model architectures for hardware compatibility
• Partner with the Infrastructure team to influence future compute cluster design

Required qualifications

• 3+ years of experience in performance engineering, site reliability engineering, or low-level software development focused on HPC or AI workloads
• Expert-level knowledge of GPU architecture
• Experience optimizing CUDA kernels or using libraries like Triton
• Deep understanding of distributed training frameworks (e.g. PyTorch Distributed, DeepSpeed, or Megatron-LM) and networking protocols
• Proficiency in profiling tools such as Nsight Systems, Nsight Compute, or Weights & Biases
• A builder mindset with the ability to work autonomously in an early-stage, fast-paced startup environment

Nice to have

• Experience working with specialized AI hardware such as TPUs or alternative accelerators
• An academic background or personal interest in Materials Science, Chemistry, or Physics
• Knowledge of MLOps practices regarding deployment of models at scale in cloud environments (AWS/GCP/Azure)
• Contributions to open-source performance optimization or machine learning libraries

Benefits

• Competitive salary
• Equity in CuspAI
• 28 days holiday (DE, NL, UK) or 21 days holiday (JP, SG, US), in addition to local public holidays
• 26 weeks parental leave for primary caregiver at full pay
• 12 weeks parental leave for secondary caregiver at full pay
• Professional development budget

About the company

CuspAI is a frontier AI company on a mission to solve the breakthrough materials needed to power human progress. The founding team is comprised of researchers in AI, chemistry and engineering.