DevOps / Deployment engineer

Responsibilities

• Deploy and configure AI systems in closed-loop (on-premise, air-gapped) environments — running on in-house GPU/server resources
• Physical setup, configuration, and support at client sites (weekly on-site visits required)
• Build and automate CI/CD pipelines
• Write custom scripts and tools tailored to specific client needs (e.g., data ingestion, integration, requirement-gathering flows) rather than relying only on pre-built tools
• Set up environments that meet server, network, and data security requirements
• Monitoring, performance tracking, and troubleshooting
• Work across multiple parallel projects at the same time

Required experience

• Real, hands-on experience deploying AI models locally/on-premise — GPU inference, clustering — this is not something we can teach; candidates must already have it
• Experience building infrastructure from scratch, not just operating within an already-built corporate environment — during the interview, be ready to explain concretely how you built something, not just which tools you used
• Linux server administration, Docker/Kubernetes, CI/CD tools (GitLab CI, Jenkins, GitHub Actions, etc.)
• Knowledge of databases and networking/security principles
• Ability to work independently — make and execute core technical decisions with minimal oversight

Personal qualities

• Outcome- and customer-problem-oriented, rather than someone who lists tool/technology names
• When faced with a new task, able to understand the underlying problem and build a solution — not just say "I use this tool for that"
• Fluent in Russian or English

Terms

• Full-time or part-time possible (depends on candidate's experience and availability)
• Part-time: infrastructure maintenance — setup, connection, configuration
• Full-time: full involvement in projects, including ownership of specific workstreams
• Weekly on-site visits to client locations required
• Compensation to be discussed based on employment format