Elevate your career with NVIDIA as an AI Performance Software Engineer. Join a hybrid team and optimize GPU kernels for large-scale AI inference systems to achieve unmatched efficiency. At NVIDIA, we're looking for seasoned engineers to architect and develop high-performance inference software.

Ideal candidates will have at least 5 years of experience and expertise in Python, C/C++, or similar languages. You'll contribute features to vLLM and benchmark GPU kernels while pushing performance boundaries in machine learning systems. Key Responsibilities:

- Architect novel frameworks for high-performance inference
- Optimize vLLM inference framework and GPU kernels
- Profile and implement cutting-edge research techniques
- Benchmark software using latest NVIDIA GPU features
- Publish original research to integrate into production software Requirements:
- PhD, Master's, or Bachelor's degree in CS or related
- 5+ years industry experience in software engineering
- Proficient in Python and C/C++, Go, or Rust
- Solid CS fundamentals in various domains
- Passionate about performance engineering in ML frameworks Bring your expertise in AI inference and performance optimization to NVIDIA’s cutting-edge team.

📌 AI Performance Software Engineer at NVIDIA (Toronto)
🏢 NVIDIA
📍 Toronto