ML Performance Engineer

ML Performance Engineer

Location: Palo Alto (Onsite)

We’re hiring an ML Performance Engineer to join a Series B AI lab valued at $1.4B that is building best-in-class image models and products (not a wrapper).

They have developed in-house foundation models at scale across thousands of GPUs and are widely recognized as having one of the strongest image models in the world . This is a product-first AI lab building creative tools currently used by millions of users .

This role focuses on large-scale training and inference optimization for in-house models, working closely with world-class research and engineering teams to maximize efficiency across GPUs and other accelerators.

What You’ll Do:

As an ML Performance Engineer, you’ll help optimize training and inference for large-scale ML models.

• Work on core training and inference systems for large models
• Improve efficiency across GPUs and other accelerators
• Profile, debug, and optimize performance bottlenecks
• Collaborate with researchers to make new models production-ready
• Improve reliability and scalability of distributed workloads
• Contribute to internal tooling and infrastructure that supports model deployment
• Evaluate and adopt new hardware and system capabilities

Bonus Points:

• Experience working on ML infrastructure, distributed systems, or performance-critical software
• Exposure to low-level optimization, GPU programming, or numerical computing
• Background in graphics, vision, or generative AI applications
• Contributions to open-source projects

Requirements:

• Strong software engineering fundamentals and problem-solving skills
• Proficiency in Python and at least one systems language (e.g., C/C++)
• Experience with machine learning frameworks (PyTorch, JAX, TensorFlow, etc.)
• Understanding of performance, scalability, or distributed computing concepts
• Interest in making systems faster and more efficient
• Bachelor’s degree in CS, Engineering, or related field (advanced degrees welcome)