Full Job Description
About the role:
We are looking for seasoned MLOps engineers with deep experience in machine learning infrastructure, large-scale model training, and production ML systems to join the team and help scale our AI capabilities across the platform.
Responsebilities:
• Design, build, and maintain scalable machine learning pipelines that support the full model lifecycle-from data ingestion and feature engineering through training, evaluation, and deployment
• Lead pretraining efforts for large transformer models on UKG's proprietary datasets, optimizing for distributed training efficiency across GPU clusters
• Develop and operate finetuning pipelines that adapt foundation models to domain-specific tasks, managing experiment tracking, hyperparameter optimization, and model versioning
• Build and maintain robust model serving infrastructure to deploy models at scale with low latency, high availability, and cost efficiency
• Implement CI/CD pipelines for ML workflows, including automated testing, validation, and canary deployment strategies for model updates
• Establish monitoring and observability for production models, including data drift detection, model performance degradation alerts, and automated retraining triggers
• Optimize training and inference costs through techniques such as mixed-precision training, model quantization, distillation, and efficient batching strategies
• Technical leadership: Serve as the technical lead on MLOps initiatives, owning end-to-end infrastructure decisions from architecture design through production operations and driving platform-wide standards
• Internal & external communication: Communicate infrastructure decisions, capacity planning, and system reliability to both internal teams (data science, engineering, leadership) and external stakeholders (customers, partners)
• Knowledge sharing & team development: Foster a culture of engineering excellence by establishing MLOps best practices, conducting architecture reviews, and mentoring team members on scalable ML systems
*** This role is a hybrid role ( 3 days in the office)***
***UKG is unable to offer sponsorship for this position.***
About you,
• Education and experience in a quantitative or engineering field. PhD plus 3 years of experience, or master's degree plus 5 years of experience, in machine learning engineering, MLOps, or a related discipline.
• Deep expertise in ML infrastructure, including distributed training frameworks, model serving platforms and orchestration tools (e.g., Kubeflow, Airflow, Prefect).
• Hands-on experience pretraining or finetuning large transformer models (1B+ parameters), including familiarity with techniques such as LoRA, QLoRA, RLHF, and DPO.
• Strong experience with cloud-based ML platforms and GPU infrastructure (e.g., Google Vertex AI, AWS SageMaker, Azure ML) and container orchestration (Kubernetes, Docker).
• Proficiency with ML experiment tracking and model registry tools (e.g., MLflow, Weights & Biases, LangSmith, Langfuse).
• Significant experience with big data tools, databases, and scalable architectures
• Advanced programming skills in Python, with expertise in ML frameworks such as PyTorch, Hugging Face Transformers, and infrastructure-as-code tools (Terraform, Pulumi, Bicep).
• Strong understanding of model optimization techniques including quantization, distillation, pruning, and efficient inference strategies.
• Proven track record of building and operating production ML systems that serve models at scale with high reliability.
• Commitment to continuous learning and staying current with the rapidly evolving ML infrastructure and foundation model landscape.
The pay range for this position is $133,900 to $190,750. The actual base pay offered may vary depending on skills, experience, job-related knowledge and work location. In addition to base pay, employees may be eligible to participate in a performance-based bonus plan and to receive restricted stock unit awards as part of total compensation. Learn more about UKG's benefits and rewards at https://www.ukg.com/about-us/careers/benefits