Engineering Lead, Inference Optimization (AI)

Location: Remote — US only

Base annual salary: $270,000–$330,000 USD

Company

hirify.global is a privacy-focused consumer AI company building a platform for individuals, applications, and AI agents.

What you will do

• Own the technical strategy for inference performance.
• Recruit and lead the Inference Optimization Team.
• Optimize GPU infrastructure across architectures including H200 and B300 GPUs.
• Improve latency, throughput, and cost per token for high-volume LLM inference workloads.
• Build reproducible benchmarking harnesses for inference engines such as vLLM and SGLang, evaluating engines, quantization schemes, and parallelism strategies.
• Optimize multivariate inference load balancing and evaluate emerging techniques, kernels, attention variants, compilation improvements, and inference hardware.

Requirements

• 8+ years of experience in performance optimization or HPC, with deep GPU architecture and parallel programming knowledge.
• 5+ years of experience leading engineering teams, combining hands-on individual contribution with team management.
• Proficiency in Python, Rust, or Go.
• Production experience with at least one high-volume LLM inference engine, such as vLLM or SGLang.
• Experience with continuous batching, PagedAttention/KV cache management, speculative decoding, quantization, CUDA graphs, and torch.compile.
• Experience with distributed inference, GPU profiling, and multi-GPU or multi-node environments.

Nice to have

• C++ and CUDA experience.
• Hands-on custom CUDA or Triton kernel development.
• Diffusion or image model inference optimization experience.
• Contributions to open-source inference frameworks.

Culture & Benefits

• Work on privacy-focused AI with zero data retention and zero training on user inputs.
• High-agency environment centered on curiosity, ownership, ethical principles, and collaboration.
• Opportunity to shape technical strategy and build an inference optimization function from the ground up.