About this role
AI Engineer (Serving) at Toss. Location: Seoul, Seoul, Korea. Role: operating frameworks, serving models, improving performance Requirements: Experience operating and tuning model-serving frameworks (vLLM, Triton, Dynamo), building stable real-time LLM/embedding serving, monitoring, automation/orchestration, and understanding GPU/LLM internals. Category: Engineering Seniority: Mid Level Tools: vLLM, Dynamo, Triton, Model Registry, GPU Commitment: Full Time Workplace: Onsite Languages: Korean