About the role 

Location: Amsterdam, Tel-Aviv
Duration: 3 months
Start date: January 2027
Compensation: Paid
Eligibility: Current University student (Computer Science or related field), Recent Graduate or Early Career specialist 
Work authorization: Permitted to work in the job's location

Capacity Management sits at the intersection of cloud infrastructure, engineering operations, and business priorities. Our team is responsible for ensuring that infrastructure capacity is available for customers, coordinating maintenance activities, and improving operational processes across multiple data centers. 

We work closely with hardware infrastructure, platform engineering, networking, and other technical teams to maintain infrastructure availability and ensure that operational issues are identified and resolved efficiently. 

We are looking for a student or recent graduate to help us improve and operate several key infrastructure management processes, focusing on GPU node repairs, node acceptance monitoring, and infrastructure maintenance coordination. 

This is an opportunity to gain hands-on experience with real-world cloud infrastructure operations. You will work with engineering teams, analyze operational metrics, track issues, and help improve processes that directly affect infrastructure availability. 

You don't need to be an infrastructure expert when you join. We value curiosity, analytical thinking, clear communication, and a willingness to understand how complex technical systems operate. 

Your responsibilities will include 

GPU Node Repair Process

Help manage and improve the node repair process across Nebius data centers to ensure infrastructure failures are identified, tracked, and resolved within agreed service level objectives. 

• Monitor GPU node availability and repair metrics across different regions and clusters. 

• Track compliance with node repair SLAs and identify cases requiring attention. 

• Analyze repair backlogs, identify nodes stuck in repair, and follow up with responsible teams. 

• Work with Hardware Infrastructure and Platform teams to investigate repair delays and improve issue resolution. 

• Maintain operational dashboards, Jira tickets, and regular status reports. 

• Identify recurring problems, process inefficiencies, and opportunities for improvement. 

Node Acceptance Monitoring

Help establish and maintain an operational monitoring process for node acceptance, ensuring that newly deployed or repaired nodes progress through acceptance testing without unnecessary delays. 

• Monitor node acceptance results across all regions and identify nodes stuck in the process. 

• Track compliance with the Node Acceptance Responsibility Agreement and associated SLAs. 

• Ensure responsible teams are aware of failed acceptance tests and outstanding actions. 

• Follow up on unresolved issues and help coordinate their resolution between engineering teams. 

• Analyze acceptance failures and identify recurring issues or bottlenecks. 

• Contribute to improving acceptance monitoring dashboards and operational procedures. 

Infrastructure Maintenance Coordination

Participate in planning and coordinating smaller infrastructure maintenance activities, such as network switch replacements, under the guidance of experienced team members. 

• Help prepare maintenance plans, timelines, and execution checklists. 

• Coordinate with participating engineering teams to ensure readiness for scheduled activities. 

• Track maintenance preparation tasks and follow up on outstanding dependencies. 

• Support maintenance communications and progress reporting. 

• Collect execution results and document outcomes, incidents, and follow-up actions. 

• Help improve existing maintenance procedures and documentation based on practical experience. 

Additional projects 

Depending on your progress and interests, you may also have the opportunity to work on larger process improvement initiatives. 

Extending the Node Repair Process 

Help expand the existing GPU node repair process to cover other infrastructure types, including CPU and worker nodes. This would involve understanding differences in their operational lifecycles, identifying relevant stakeholders, and adapting existing monitoring, ticketing, and SLA processes. 

Improving Major Maintenance Documentation 

Help review and restructure the existing Major Maintenance documentation according to established documentation guidelines. The goal would be to make the process easier to understand and follow, with clear responsibilities, instructions, templates, and references for participating teams. 

What we expect 

• You are currently completing, or have recently completed, a degree in engineering, computer science, information systems, operations management, or a related field. 

• You have strong analytical and problem-solving skills and are comfortable working with data and operational metrics. 

• You can organize information, track multiple activities, and follow up on outstanding tasks. 

• You communicate clearly and are comfortable collaborating with technical specialists and managers. 

• You are proactive in identifying problems and investigating their causes rather than simply reporting them. 

• You can work with established processes while also identifying opportunities to improve them. 

• You pay attention to detail and take responsibility for completing assigned tasks. 

• You are interested in cloud infrastructure, technical operations, and cross-functional project management. 

You do not need previous experience operating large-scale cloud infrastructure. Curiosity, initiative, and the ability to learn quickly are more important. 

It would be an advantage if you have 

• Basic knowledge of cloud infrastructure, servers, networking, or data center operations. 

• Experience working with monitoring or data visualization tools such as Grafana. 

• Familiarity with SQL or other data analysis tools. 

• Experience using Jira, Confluence, or similar collaboration tools. 

• Experience coordinating university, volunteer, or personal projects involving multiple participants. 

• An interest in infrastructure operations, reliability engineering, or technical project management. 

What you will gain 

This Early Talent program provides hands-on experience with the operational challenges of running large-scale AI cloud infrastructure. 

You will have the opportunity to: 

• Understand how GPU clusters and cloud infrastructure are operated and maintained. 

• Learn how infrastructure failures are detected, investigated, and resolved. 

• Gain practical experience with operational monitoring, SLA management, and incident prevention. 

• Work with engineers across hardware infrastructure, compute, networking, and platform teams. 

• Participate in infrastructure maintenance planning and execution. 

• Develop analytical, communication, and technical project coordination skills. 

• Contribute to processes that directly impact infrastructure availability and customer experience. 

By the end of the Early Talent Program, you should have a practical understanding of how infrastructure operations are coordinated across engineering teams, along with experience identifying operational problems and driving them toward resolution. 

What success looks like 

A successful Early Talent will result in: 

• Regular monitoring and reporting of GPU node repair performance, with SLA violations and outstanding issues clearly identified. 

• Improved visibility into node repair backlogs and more consistent follow-up on delayed repairs. 

• An established routine for monitoring node acceptance, tracking SLA compliance, and ensuring responsible teams take action on failed tests. 

• Successful coordination of one or more smaller infrastructure maintenance activities under supervision. 

• Improvements to operational dashboards, reporting, or process documentation. 

• Demonstrated ability to collaborate with technical teams, investigate issues, and independently manage assigned operational tasks. 

If the Early Talent Program is successful, we will be happy to discuss an opportunity to continue with the team in a full-time position, subject to performance, mutual fit, and business needs.