About the role:
Join our dynamic team in Berlin as a Senior MLOps Engineer and play a crucial role in shaping the future of our ML infrastructure. You'll be responsible for designing, building, and maintaining scalable and resilient machine learning pipelines that support real-time and batch model deployment and monitoring. Your expertise will be key in supporting various AI and machine learning initiatives across the organization, making a real-time, real-life impact on our business.
What you will be working on:
ML Platform Development and Maintenance
- Design and implement a scalable and reliable ML platform using tools such as MLflow, Kubeflow, and Metaflow
- Ensure timely and reliable delivery of ML model predictions, with continuous monitoring for performance, accuracy, and reliability
- Maintain and extend our in-house batch and real-time feature platforms powering numerous feature pipelines
Cross-functional Collaboration
- Work closely with data scientists and data engineers to understand model and feature requirements
- Translate requirements into scalable and reliable pipelines
- Collaborate on deploying, monitoring, and optimizing ML models in production
Continuous Improvement and Innovation
- Actively contribute to the team's knowledge sharing and participate in tech talks
- Drive architectural decisions as the subject matter expert (SME) in MLOps
- Stay abreast of industry trends and advancements in MLOps technologies to improve our ML Platform
Who we are looking for:
- Extensive experience with machine learning workflows and MLOps tools such as Metaflow and MLflow
- Strong knowledge of model deployment, serving, and monitoring using tools like AWS SageMaker, KServe, Seldon Core
- Proficiency in designing and implementing model deployment pipelines using Github Actions, ArgoCD, or similar tools
- Experience with orchestration tools such as Airflow, ArgoCD, Prefect
- Familiarity with data versioning, data governance practices, and model versioning
- Understanding of reproducibility and experimentation using tools like MLflow or Weights & Biases
- Knowledge of model training and inference using popular frameworks like Scikit-learn (TensorFlow or PyTorch is a plus)
- Excellent collaboration skills, particularly in working with data scientists on model deployment and optimization
