LLM Inference Engineer

Responsibilities

• Optimize inference performance for large language models and design improvements
• Develop serving engines to deploy the EXAONE model stably across various frameworks
• Contribute to open-source LLM serving frameworks such as vLLM, SGLang, and Ollama
• Apply and improve chat_template to ensure correct LLM behavior across diverse languages and environments
• Design NPU/GPU-based serving strategies and implement model inference acceleration algorithms

Requirements

• Experience in model inference acceleration and inference back-end implementation
• Experience designing and implementing NPU/GPU-based model serving strategies
• Contribution experience to open-source LLM serving frameworks (vLLM, SGLang, Ollama)

Preferred Qualifications

• Understanding of the machine learning model lifecycle (development, training, serving, deployment)
• Proficiency in AI/ML technologies such as Python, C++, TensorFlow, PyTorch, ONNX, CUDA, and NPU
• Deep understanding of Transformer-based models and LLM architecture
• Knowledge of LLM decoding strategies (Greedy Search, Beam Search, Top-k Sampling)
• Experience designing and applying chat_template across diverse environments

About the Company

LG AI Research Institute optimizes the AI model lifecycle from development to deployment through MLOps pipelines. The institute also strengthens security, manages infrastructure, and optimizes resources to support stable AI service operations.