Remote job
Senior MLOps muhandis
Job details
About this role
Role overview Build and operate machine learning systems on customer GPU clusters, making production model serving reliable and observable. The work spans deployment, resource scheduling, data and model pipelines, and monitoring in on-premise environments.
Responsibilities - Deploy and serve models on GPU infrastructure, using tools such as vLLM, Triton, and Kubernetes. - Support data, training, and evaluation pipelines, including model version management. - Monitor latency, operating cost, and model quality in on-premise deployments. - Help deliver private large language model and retrieval-augmented generation platforms for enterprise customers.
Requirements - At least five years of MLOps or machine learning platform engineering experience. - Practical experience with model serving and GPU scheduling. - Experience building or operating data, training, and evaluation pipelines and managing model versions. - Ability to monitor production workloads in on-premise environments. - Current authorization to live in South Korea and working proficiency in English; Korean is an advantage.
Benefits and work setup - Full-time employment and market-level compensation based on experience. - Work with a small, senior team and direct access to executive leadership.