MLOps/AIOps Engineer 

Location: Casablanca (onsite work mode) 

We are looking for a skilled MLOps/AIOps Engineer to join our team. In this role, you will bridge the gap between machine learning model development and production deployment. You will work with cross-functional teams (data scientists, software engineers, DevOps engineers) to ensure scalable, efficient, and robust machine learning models are delivered into our environments.

 

What you’ll do

Automate and streamline the deployment of machine learning models into production systems. Ensure that ML models are properly integrated with applications, services, and infrastructure. Build and maintain CI/CD pipelines for machine learning models, enabling rapid experimentation and iteration while ensuring quality and performance in production. Leverage priority and open-source technologies to support CI/CD pipelines. Establish and maintain robust monitoring systems to track model performance in production, ensuring continuous evaluation and early detection of issues. Implement real-time performance metrics to evaluate key indicators such as accuracy, latency, and resource usage, ensuring models meet business objectives and user needs. Set up monitoring for model performance in production and address any issues that arise. This includes performance degradation, model drift, and other production challenges. Work closely with data scientists and architects to understand model requirements and deployment constraints. Collaborate with DevOps, software engineers, and other stakeholders to ensure a seamless transition from model development to production. Design and manage infrastructure for training and serving models at scale. This might include cloud resources (AWS, Azure, GCP, OCI), containerization (Docker, Rancher, Kubernetes), and orchestration tools. Build robust data pipelines for training and testing models. Automate the entire machine learning lifecycle from data preprocessing to model serving and monitoring. Ensure that models, datasets, and experiments are versioned and reproducible. Implement version control for models and maintain an effective model registry.

Qualifications

2+ years of experience in MLOps, DevOps, or software engineering, with at least one year focused specifically on machine learning model deployment. Proficiency in Python (most common for ML workflows) and experience with languages such as Java, or Go is a plus. Familiarity with popular machine learning frameworks (TensorFlow, PyTorch, Scikit-learn) and tools such as MLflow, Kubeflow, or TFX. Experience working with cloud platforms like AWS, GCP, Azure or OCI. Proficiency in containerization technologies (Docker, Rancher) and orchestration (Kubernetes, Helm). Experience with CI/CD tools (e.g., Jenkins, GitLab CI,  GitHub Actions) and Git. Familiarity with tools for model monitoring, logging, and observability (e.g., Prometheus, Grafana,). Strong understanding of data pipeline design, and data storage technologies (e.g., SQL, Object Storage). Strong communication skills to work with cross-functional teams . Familiarity with distributed systems and big data frameworks like Hadoop, Spark, etc. Experience in automating machine learning (ONNX) or model training optimization. Experience in model interpretability, explainability, and fairness.

Career Level - IC3