RoleCerebras is looking for an AI Data Center Infrastructure Engineer to turn cluster architecture into precise, deployment-ready infrastructure designs.
You will own the critical step between a high-level system design and its implementation in a real data center. This includes defining rack layouts, equipment placement, bills of materials, power requirements, and port and cable mappings for compute, network, and storage systems.
Every deployment presents different constraints. You will adapt reference designs to the available space, power, cooling, connectivity, hardware, and logistics while preserving system performance, reliability, scalability, and serviceability.
You will also help automate this work. By building tools, validation checks, and reusable design frameworks, you will make each deployment faster, more accurate, and easier to operate at scale.
This role is well suited to an engineer who enjoys detailed systems work, practical problem-solving, and collaborating across hardware, networking, software, and data center operations.
Responsibilities - Translate cluster architectures and technical requirements into complete, deployment-ready infrastructure designs.
- Produce and maintain rack elevations, equipment layouts, bills of materials, power allocations, port maps, and cable maps.
- Adapt reference designs to site-specific constraints involving space, power, cooling, network connectivity, logistics, and hardware availability.
- Validate that compute, storage, networking, power, and cabling designs work together as an integrated system.
- Evaluate design trade-offs and recommend practical solutions that balance performance, reliability, scalability, cost, serviceability, and deployment schedule.
- Partner with systems, network, hardware, manufacturing, supply-chain, and data center operations teams throughout the design and deployment process.
- Participate in design and deployment-readiness reviews, identifying risks, inconsistencies, and missing requirements before installation begins.
- Support deployment teams by resolving physical integration issues and documenting approved design changes.
- Contribute to root-cause analysis when deployment or integration problems occur, and incorporate lessons learned into future designs.
- Develop scripts and tools that automate design generation, BOM creation, cable mapping, consistency checks, and other repeatable workflows.
- Improve templates, standards, documentation, and design processes so deployments become increasingly repeatable and scalable.
Qualifications - A bachelor's or master's degree in computer engineering, electrical engineering, computer science, or a related discipline-or equivalent practical experience.
- One or more years of relevant experience in infrastructure engineering, data center design, systems integration, hardware deployment, or a related field.
- Experience creating or working with rack elevations, equipment layouts, bills of materials, port maps, or cable maps.
- A practical understanding of server, storage, and networking hardware and how these systems are physically integrated.
- Familiarity with data center power, cooling, space, cabling, and serviceability constraints.
- Experience using Python, Bash, PowerShell, or another language to automate technical workflows.
- Strong analytical skills and close attention to detail.
- The ability to turn incomplete or changing requirements into clear, actionable designs.
- Clear written and verbal communication skills, including the ability to collaborate across several engineering disciplines.
Helpful ExperienceThe following experience is valuable but not required:- Designing or deploying high-density AI, HPC, or large-scale compute clusters.
- Working with high-speed Ethernet or InfiniBand networks, optical transceivers, fiber, DACs, or structured cabling.
- Using infrastructure documentation, diagramming, DCIM, CAD, or source-controlled design tools.
- Developing automated validation, configuration-generation, or infrastructure-modeling tools.
- Supporting data center deployment, commissioning, hardware integration, or deployment-related incident analysis.
- Working with contract manufacturers, colocation providers, installation partners, or supply-chain teams.
