AI DevOps (Remote)

View full description

Requirements:

• Proven experience as a DevOps Engineer, preferably with exposure to AI/ML workflows and production environments.
• Strong knowledge of CI/CD pipelines and automated deployment best practices in cloud environments.
• Hands-on experience with containerization and orchestration tools, particularly Docker and Kubernetes.
• Proficiency in cloud infrastructure management with an emphasis on optimizing resource usage and costs.
• Expertise in monitoring tools such as Prometheus, Grafana, or the ELK stack.
• Proficiency in scripting and automation using Python, Bash, or similar languages.
• Deep understanding of version control systems and branching strategies (e.g., Git).
• Experience with MLOps frameworks like MLflow or Kubeflow for model lifecycle management.
• Knowledge of security best practices in DevOps and cloud environments.
• Strong communication skills for collaboration with cross-functional teams and stakeholder alignment.
• Problem-solving mindset with a proactive approach to optimizing operations.

Responsibilities:

• Design, implement, and manage scalable CI/CD pipelines for the deployment and lifecycle management of AI models.
• Conduct technical investigations of AI models, focusing on evaluating performance, scalability, and cost-efficiency.
• Automate the deployment, monitoring, and maintenance of AI applications in production environments.
• Continuously improve workflows by implementing automation and monitoring best practices.
• Optimize infrastructure usage and manage costs while ensuring effective cloud resource utilization.
• Establish security best practices for AI workloads, including data encryption and access controls.
• Collaborate with cross-functional teams to ensure seamless AI model integration into production workflows.
• Take ownership of AI model demonstrations during internal sprints and prepare performance showcases.
• Develop logging and monitoring frameworks to ensure reliability and uptime of AI services.
• Participate in incident response processes and troubleshoot issues in AI pipelines.