About the Team
The Risk team develops automated IT solutions for risk management, creating models and running real-time production inference under specified SLAs while integrating external services and making data-driven decisions. The Data Science team develops scoring models, including neural network-based approaches.
Responsibilities
•Evaluate Feature Store / Feature Registry solutions, prepare a recommendation, and design the architecture and feature lifecycle process from experiment to stable production
•Lead implementation by engineering and platform teams
•Design, build, and own ML training and deployment pipelines, including experiment tracking, model registry, model CI/CD, packaging, and handoff to production
•Select a platform such as MLflow or alternatives, establish versioning and validation standards, and continuously evolve the infrastructure
•Select tooling and set up monitoring for model quality and feature health in collaboration with the DS team
•Define the alerting and response process
Requirements
•3+ years of experience in ML Engineering, Data Engineering, or DevOps, including hands-on deployment and maintenance of production ML systems
•Experience building ML training pipelines with experiment tracking and a model registry such as MLflow or W&B
•Understanding of feature store concepts and train-serve consistency challenges
•Strong Python skills and sufficient understanding of ML frameworks to package, serve, and debug models
•Experience with Docker and ML pipeline orchestration tools such as Kubeflow, Argo Workflows, or Metaflow
•Solid SQL skills and understanding of data warehouse architecture
•English B1 or higher
Conditions
•Relocation support to one of the hubs in Cyprus, Serbia, Georgia, or Kazakhstan, including assistance for the employee and their family
•Healthcare coverage
•Education budget for language lessons, professional training, and certifications
•Wellness budget for mental health and fitness activity reimbursements
•20 days of annual leave and paid sick leave
