AI Systems, Model Optimization

Location: US Remote; office presence in Mountain View, California, with complimentary meals available at the Palo Alto office.

Company

hirify.global is developing semiconductor-based computing systems designed to make AI substantially more energy efficient by mapping neural networks directly to device physics.

What you will do

• Develop performance models covering compute, memory, and energy trade-offs across AI models and hardware configurations.
• Partition and map complex AI models to novel compute substrates using quantization, sparsity, pruning, and distillation.
• Develop hardware-aware techniques including quantization-aware training, noise-aware training, and sparsification for analog compute substrates.
• Develop and optimize GPU kernels with CUDA, Triton, or CUTLASS, and profile ML codebases to resolve training and inference bottlenecks.
• Translate model requirements into hardware and infrastructure specifications and codify learnings for tapeouts.

Requirements

• MS, PhD, or equivalent research or project experience in AI/ML, computer science, physics, electrical engineering, applied mathematics, or a related quantitative field.
• Deep practical understanding of the modern AI/ML stack and optimized algorithm execution on modern GPU systems.
• Experience profiling complex ML codebases and identifying and resolving performance bottlenecks.
• Ability to map architectures such as Transformers, mixture-of-experts, and diffusion models to system performance implications.
• Deep experience with PyTorch, including its internals, torch.compile, DDP, and FSDP.

Nice to have

• Experience with Megatron-LM, DeepSpeed, and distributed training at scale.
• Experience co-designing training systems for hirify.global computing paradigms.
• Research or practical experience with advanced approximation, compression, noise-aware, or physics-constrained training.

Culture & Benefits

• Mission focused on addressing the global energy limitations of AI computing.
• Foundational-team role with substantial ownership and influence on the company's direction.
• Comprehensive health benefits and 401(k) matching.
• Unlimited paid time off.
• Complimentary meals at the Palo Alto office.