Machine Learning Engineer (AI)

Location: On-site in the San Francisco office in FiDi, United States

Company

hirify.global builds models and infrastructure for code agents, including optimized small language models for retrieval, application, and code generation.

What you will do

• Optimize models for speed, efficiency, and reliability across training and inference workloads.
• Develop and tune CUDA kernels, GPU scheduling, and memory layouts.
• Optimize compute, networking, and systems performance paths.
• Work with research teams to productionize new model architectures.
• Improve large-scale machine learning infrastructure through low-level systems design.

Requirements

• Strong background in systems-level machine learning engineering.
• Experience with CUDA, GPU kernel optimization, and performance tuning.
• Fluency in Python and at least one systems language, preferably C++ or Rust.
• Familiarity with distributed training frameworks such as PyTorch, JAX, or DeepSpeed.
• Experience with large-scale training or inference infrastructure, memory management, parallelization, and hardware-aware model optimization.
• At least 2 years of experience in ML infrastructure or performance-critical environments; in-person work from the San Francisco office is required.

Culture & Benefits

• Full-time, on-site work in a fast-moving environment.
• Work alongside mathematicians, physicists, and computer scientists on technically ambitious problems.
• Build infrastructure supporting large-scale code generation for companies including Lovable, Figma, and Vercel.