Senior MLOps Engineer

clariticloudinc; United States; Remote; USD 105,000 - 140,000; Full time

Job summary

clariticloudinc is seeking the first MLOps engineer to build production-ready ML infrastructure, spanning pipelines, cloud compute, model serving, and automation. You will work closely with AI researchers and engineers to create reproducible, scalable, observable, and cost-conscious systems.

The role covers end-to-end pipelines, model training and serving, CI/CD, monitoring for drift and data quality, and cost-optimization of training workloads in a remote setup with occasional in-person travel.

Qualifications

• Six or more years in software or machine-learning engineering, including at least three years in MLOps or ML infrastructure.
• Strong programming experience with Python, C/C++, or Bash.
• Demonstrated experience deploying and operating production ML models.

Responsibilities

• Build and maintain end-to-end pipelines for model training, evaluation, deployment, and agent-based AI workflows.
• Develop and optimize infrastructure for training and serving models across cloud providers and compute environments.
• Create tools for data pipelines, labeling, experiment and performance tracking, and automated retraining.
• Help research and engineering teams productionize prototypes with reproducible, scalable workflows.
• Establish CI/CD practices, automated testing, and deployment processes for ML systems.
• Monitor production model health, performance, drift, and data quality; tune performance and training costs.

Skills

Python

C/C++

Bash

Docker

Kubernetes

MLflow

Weights & Biases

Kubeflow

CI/CD

Experiment tracking

Tools

Docker

Kubernetes

MLflow

Weights & Biases

Kubeflow

Job description

Role overview

As the first MLOps engineer, you will build the infrastructure that moves machine-learning models and AI workflows from research into dependable production use. The role spans pipeline design, cloud and compute infrastructure, model serving, automation, and ongoing monitoring. You will work closely with AI researchers, machine-learning engineers, and broader engineering teams to make systems reproducible, scalable, observable, and cost-conscious.

Responsibilities

• Build and maintain end-to-end pipelines for model training, evaluation, deployment, and agent-based AI workflows.
• Develop and optimize infrastructure for training and serving models across cloud providers and compute environments.
• Create tools for data pipelines, labeling, experiment and performance tracking, and automated retraining.
• Help research and engineering teams productionize prototypes with reproducible, scalable workflows.
• Establish CI/CD practices, automated testing, and deployment processes for machine-learning systems.
• Monitor production model health, performance, drift, and data quality; tune performance and training costs.

Requirements

• Six or more years in software or machine-learning engineering, including at least three years in MLOps or ML infrastructure.
• Strong programming experience with languages such as Python, C/C++, or Bash.
• Demonstrated experience deploying and operating production machine-learning models.
• Experience designing scalable systems with Docker and Kubernetes, and using a major cloud platform with GPU orchestration.
• Familiarity with CI/CD tools and experiment tracking or pipeline systems such as MLflow, Weights & Biases, or Kubeflow.
• Understanding of data versioning, model reproducibility, and production monitoring; strong problem-solving and collaboration skills.

Nice to have

• Experience training models from scratch, model optimization methods such as quantization or distillation, infrastructure-as-code, distributed systems, high-performance computing, or open-source contributions.

Benefits and work setup

The stated base salary range is CAD $150,000–$200,000, based on skills and experience. The organization describes competitive compensation, time off, and benefits, which may vary by employment type. The role is remote, with occasional in-person travel, typically once or twice per year.