The IT teams at PSB bank and its subsidiary IT company PSB Lab already have over 4,000 employees. We develop and implement fintech products and applications for clients, support complex IT systems of the bank, and are among the top three digital banks in Russia.
In the Information Security Block, within the Information Protection Coordination Department of the Security Directorate, there is an opening for a Middle DevOps Engineer (ML/AI, DevSecOps, MLSecOps).
Key Tasks:
- Design and support of infrastructure for training and inference of ML and LLM models.
- Automation of CI/CD pipelines for software development and Data Science teams: software assembly, testing, image building, model testing, deployment to various environments.
- Work with GPU infrastructure and management of on-premise GPU servers: OS driver configuration, monitoring, resource utilization optimization.
- Implementation of DevSecOps practices: container vulnerability scanning (Trivy), SAST/DAST, secrets management (HashiCorp Vault), network security policies.
- Ensuring the security of ML and AI pipelines (MLSecOps): data protection, access control to models, monitoring of drift and anomalies.
- Monitoring and logging of infrastructure and applications (Prometheus, Grafana).
What is important to us:
- Experience in administration and confident proficiency in Linux (AstraLinux).
- Scripting skills for automation (bash, python).
- Experience in configuring and supporting DBMS (Postgres, ClickHouse).
- Confident proficiency in Docker, experience in building CI/CD (GitLab CI / Jenkins / GitHub Actions).
- Understanding of security fundamentals: access control, network policies, principle of least privilege.
- Experience with GPU instances, experience with monitoring systems (Prometheus, Grafana).
- Basic understanding of the ML model lifecycle: difference between training and inference, working with frameworks (PyTorch/TensorFlow).
- Understanding of the main elements of AI infrastructure (LLM infrastructure, AI agents, MCP servers).
- Willingness to delve into the requirements of AI engineers and ML developers (Data Science) and propose infrastructure solutions.
Will be an advantage:
- Experience in configuring Kubernetes clusters (helm, operators, writing manifests).
- Experience with configuration management systems and familiarity with Infrastructure as Code (Ansible, Terraform, etc.).
- Experience in deploying and maintaining LLMs (Large Language Models) in production environments (vLLM, HuggingFace).
- Monitoring LLM performance.
- Knowledge of MLOps tools: MLflow, DVC.
- Practical experience in implementing DevSecOps/MLSecOps.
What we offer:
- Official employment in accordance with the Labor Code of the Russian Federation.
- Work schedule: 5/2 from 9:00 to 18:00, shortened working day on Fridays until 16:45.
- Hybrid work format after the probationary period (1 day per week remote).
- Competitive income level.
- Additional payment for vacation and sick leave.
- "Benefits cafeteria": Voluntary medical insurance for the employee and family members, reimbursement of vacation expenses, sports services, purchases on the marketplace "PSB Market".
- Additional benefits upon marriage and childbirth.
- Financial support in certain life situations.
- Free employee support program: legal, financial, and psychological consultations, assistance with household matters, car assistance, corporate discounts, career guidance for employees' children.
- Opportunities for professional development and internal and external professional training.
- Corporate matching pension program.
