Machine Learning Engineer (LLM)

Location: Yerevan, Armenia

Company

hirify.global develops production AI solutions and infrastructure for deploying and operating machine learning systems.

What you will do

• Design, deploy, and optimize large language model inference systems for production.
• Evaluate serving performance across latency, throughput, memory utilization, and cost-efficiency.
• Implement model-serving solutions with vLLM, TensorRT-LLM, SGLang, Triton Inference Server, or similar technologies.
• Develop model conversion, deployment, benchmarking, and evaluation workflows.
• Apply quantization techniques and profile GPU workloads to identify and resolve performance bottlenecks.
• Collaborate with Platform, Infrastructure, and Product Engineering teams on scalable, reliable AI services and architecture decisions.

Requirements

• 3+ years of ML engineering experience focused on model deployment, serving, and inference optimization.
• Production experience with modern LLM serving frameworks such as vLLM, TensorRT-LLM, SGLang, or similar technologies.
• Deep understanding of transformer architectures, attention mechanisms, KV caching, autoregressive generation, and prefill versus decode phases.
• Experience with FP8, INT8, AWQ, GPTQ, or similar quantization techniques and their accuracy, latency, and hardware trade-offs.
• Advanced Python skills, strong performance profiling experience, and working knowledge of PyTorch, ONNX, TorchScript, and TensorRT.
• Ability to collaborate with platform, infrastructure, and product teams in production environments.

Nice to have

• Knowledge of GPU architecture, HBM bandwidth, tensor cores, NVLink, and modern NVIDIA accelerators such as B200, H100, or L40S.
• Experience with Kubernetes, containerized ML workloads, and Triton Inference Server.
• Knowledge of distributed inference, tensor parallelism, NCCL, RDMA/InfiniBand, and GPUDirect.
• Experience with speculative decoding, prefix caching, multi-tenant serving, or LoRA adapter hot-swapping.

Culture & Benefits

• Full-time role focused on production AI infrastructure.
• Cross-functional collaboration with platform, infrastructure, and product engineering teams.
• Opportunity to evaluate emerging LLM serving, inference optimization, and AI infrastructure technologies.