AI/ML Solution Architect (Distributed Training)

Location: Remote from anywhere in the U.S. or Canada

Total compensation up to ~$500k, including base and variable compensation, depending on level and experience.

Company

hirify.global supports a fast-moving AI infrastructure team working on large-scale machine learning workloads and GPU platforms.

What you will do

• Design and validate production-grade distributed training and large-scale inference architectures on clusters ranging from tens to thousands of GPUs.
• Debug, optimize, and scale customer ML workloads across multi-node GPU environments.
• Provide technical authority on GPU performance, networking, and schedulers, making trade-offs at scale.
• Translate customer needs into concrete platform requirements and collaborate with engineering, product, and R&D on roadmap decisions.
• Work hands-on in customer environments rather than only providing high-level advice.

Requirements

• Hands-on experience designing and operating production-grade, multi-node GPU workloads for training or inference.
• Strong background in distributed deep learning, including PyTorch Distributed and DeepSpeed, on GPU clusters.
• Deep understanding of GPU architecture and interconnects, including H100/A100-class GPUs, NVLink, and InfiniBand.
• Experience with Kubernetes or Slurm and performance tuning using GPU profiling and monitoring tools.
• Experience limited to single-node training, high-level AI strategy, or non-production research environments is not suitable for this role.

Culture & Benefits

• Hands-on technical role within an AI-focused customer solutions team.
• Collaboration with senior AI and infrastructure engineers on large-scale GPU platforms.
• Direct exposure to modern GPUs such as H100 and B200 and ML frameworks including PyTorch and, in some environments, JAX.
• Remote work available from anywhere in the U.S. or Canada.