MLOps Engineer (DevOps, Python)

Job description

About the Role

We are seeking an MLOps Engineer to join our team and streamline the deployment, monitoring, and lifecycle management of our machine learning models. The ideal candidate will work closely with Data Scientists, Data Engineers, and DevOps teams to develop and maintain ML pipelines, optimize model performance, and ensure scalable deployment in production environments. This role requires strong problem-solving skills, hands-on experience in cloud-based ML operations, and a passion for automation and efficiency.

Key Responsibilities

• Data Model Deployment & Automation:
  • Design, implement, and automate data pipelines to deploy models into production.
  • Ensure seamless integration of ML models into cloud environments.
• Monitoring & Maintenance:
  • Monitor deployed models for drift, accuracy, and performance degradation.
  • Retrain and update models as needed based on real-time data feedback.
• CI/CD for ML:
  • Develop and maintain CI/CD pipelines for model training, testing, and deployment, providing scheduled visibility for QC checks.
  • Implement version control for datasets and models to track changes effectively.
• Infrastructure & Scalability:
  • Optimize cloud infrastructure for cost-efficient ML model serving and storage.
  • Service integration to optimize infrastructure utilization.
  • Work with Kubernetes, Docker, and cloud services (AWS, GCP, or Azure) to ensure scalability.
• Collaboration & Compliance:
  • Work closely with Data Scientists and Engineers to improve ML workflows.
  • Work with external customers to ensure smooth integration and customer experience.
  • Ensure ML pipelines comply with data security and governance policies.

OKRs (Objectives and Key Results)

Objective 1: Efficient, Reliable and Accurate Scaled Data Model Deployment

• Key Result 1.1: Deploy new ML models within X days of final model approval and Y accuracy.
• Key Result 1.2: Maintain 99% uptime of deployed ML models.

Objective 2: Continuous Monitoring & Optimization

• Key Result 2.1: Implement automated monitoring for model performance (meeting time threshold) and data drift.
• Key Result 2.2: Maximize infrastructure utilization and minimize infrastructure costs. 
• Key Result 2.3: Streamline CI/CD workflows to accelerate model deployment by X% and maintain full traceability of models and datasets.

Objective 3: Streamlined Data Infrastructure

• Key Result 3.1: Smoothly integrate with front-end and customer data source
• Key Result 3.2: Scale up and ensure load balancer for customer experience.

Your skills and experience

About You

• Experience: 2-5 years in ML Ops, DevOps, Data Engineering, or Cloud AI/ML pipeline roles
• Preferred Industry Background: AI-driven startups, cloud computing, data intelligence firms or consulting
• Attention to Detail: Ability to detect errors and performance issues in ML pipelines.
• Time Management: Capability to manage multiple models and deployment cycles efficiently.
• Problem-Solving Skills: Ability to troubleshoot infrastructure, deployment, and model performance issues.
• Collaboration: Strong team player who works well with data scientists and engineers; works well with external customers.
• Adaptability: Willingness to learn new ML frameworks and deployment strategies.

Qualifications 

• Bachelor's/Master’s degree in Computer Science, Data Science, or Software Engineer.
• Experience with MLOps tools such as Triton Server, Grafana, Prometheus, MLflow, Airflow, Kubeflow, or SageMaker.
• Proficiency in Python, TensorFlow/PyTorch, and cloud computing platforms (AWS, GCP, Azure).
• Experience with CI/CD tools like Jenkins, GitHub Actions, or GitLab CI/CD.
• Familiarity with containerization and orchestration (Docker, Kubernetes).
• Strong understanding of model monitoring, logging, and retraining workflows.

Why you'll love working here

Compensation and Benefits

• Salary: Competitive, based on performance.
• Performance Bonus: Available for exceeding key objectives.
• Career Growth: Opportunities for promotion and specialization in MLOps, Data Engineering or Product Manager.
• Working Hours: Flexible arrangement, a minimum commitment of 40 hours per week