описание
Yuno builds payment infrastructure that connects companies to more than 300 payment methods worldwide through a single API. It uses AI and modern technology for intelligent transaction routing and fraud prevention across more than 80 countries.
задачи
- Design, build, and maintain the MLOps platform, including experiment tracking, model registry, versioning, and reproducible training pipelines
- Establish CI/CD practices for ML, with automated testing, validation gates, and promotion workflows from development to production
- Define standards and tooling for feature stores, model artifacts, and reproducible environments across teams
- Take models from research or prototype stages to robust, scalable production services
- Build low-latency, highly available serving infrastructure for batch, online, and real-time inference
- Implement monitoring for model performance, data drift, and concept drift, with alerting and rollback paths
- Partner with data science teams to prepare models for production constraints such as latency, cost, and scale
- Automate model retraining, evaluation, and deployment pipelines
- Build self-healing and automatic rollback mechanisms triggered by performance or drift thresholds
- Create tools that let ML practitioners ship models without deep infrastructure expertise
- Integrate ML models with streaming data platforms for real-time feature computation and inference
- Design low-latency feature pipelines connecting batch and streaming data sources
- Ensure consistency between offline training and online serving feature computation
- Design and integrate agentic workflows, including LLM-based agents and tool-calling pipelines, alongside traditional ML models
- Build observability, guardrails, and evaluation frameworks for reliable production agentic systems
- Explore how agents can automate parts of the ML lifecycle, including monitoring, triage, and retraining decisions
требования
- 5 To 8 years of experience in ML engineering, MLOps, or backend infrastructure with ML systems in production
- Strong software engineering fundamentals and ability to own services end to end
- Experience with model serving frameworks such as Seldon, KServe, BentoML, or TorchServe, and orchestration tools such as Airflow, Kubeflow, or MLflow
- Hands-on experience with streaming systems such as Kafka, Kinesis, or Flink
- Familiarity with Docker and Kubernetes
- Experience with observability tooling, including metrics, tracing, and logging, for ML or distributed systems
- Strong communication skills and comfort working cross-functionally with data science, platform, and product teams
- Fluent English
- Based in Europe
- Будет плюсом: exposure to LLM and agent frameworks and evaluation practices, experience in regulated or high-throughput domains such as fintech, payments, or healthcare, contributions to open-source MLOps or agentic tooling, experience with cloud ML platforms such as SageMaker, Vertex AI, or Databricks
условия
- Competitive compensation
- One-time home office setup allowance
- Work equipment
- Stock options
- Health plan wherever you are
- Flexible days off
- Language, professional, and personal growth courses