Software Engineer – AI Infrastructure & Performance

Seeking for a software engineer to join a team building high-performance AI infrastructure. In this role, you'll work on tools that extract, execute, and analyze AI/ML operators across various hardware platforms. You’ll develop systems for profiling, collecting, storing, and querying performance data from large-scale AI workloads. Your work will directly impact how next-generation AI models are deployed and optimized across platforms.

Responsibilities:

• Extract and run AI operators (e.g. PyTorch, Triton) on various hardware
• Profile performance and collect metrics (latency, memory, etc.)
• Design and maintain scalable databases to store and query performance data
• Build programmatic and web interfaces for analysis
• Integrate tools with CI pipelines and testing frameworks
• Collaborate with cross-functional teams on AI model deployment and evaluation

Required Skills:

• Strong Python programming experience
• Proficiency with PyTorch, CUDA, and Triton kernels
• Experience with performance profiling tools (e.g., Kineto, dispatcher)
• Database and SQL proficiency
• Familiarity with Linux and Bash scripting
• Exposure to LLMs (e.g., LLaMA or similar)
• Experience with CI/CD and test automation workflows