Infrastructure Design Engineer

Together AI Together AI · Data AI · San Francisco, CA · Business Operations

This role focuses on the physical infrastructure design of data centers that house AI GPU clusters. Responsibilities include designing whitespace layouts, power distribution, cooling, and structured cabling to support high-density AI hardware. The role requires expertise in data center design, critical facilities engineering, and collaboration with various engineering and operational teams.

What you'd actually do

  1. Architect HPC clusters by designing whitespace layouts, including rack placement, aisle configuration, hot/cold aisle containment, equipment density, and airflow strategy for high-density GPU deployments
  2. Collaborate with electrical and mechanical engineers to integrate power and cooling infrastructure into whitespace environments
  3. Collaborate with Network Engineering to define and validate physical layer requirements (structured cabling, pathway planning, port density) for high-speed AI cluster interconnects, ensuring design compatibility with both physical and logical network architectures.
  4. Advise Data Center build teams/ contractors to ensure data center build out matches design and architecture specifications. Provide direction to optimize performance, scalability and cost-effectiveness
  5. Develop and maintain CAD/BIM drawings, schematics, capacity planning models, and technical documentation to support site design, construction, operations, and audits of data center white space

Skills

Required

  • data center design
  • critical facilities engineering
  • infrastructure delivery
  • power distribution
  • cooling strategy
  • structured cabling
  • rack layouts
  • CAD/BIM tools (AutoCAD, Revit)
  • capacity planning
  • TIA-942
  • Uptime Institute Tier guidelines
  • ASHRAE thermal recommendations
  • local codes and safety regulations
  • MEP engineers
  • general contractors
  • equipment vendors

Nice to have

  • liquid cooling systems
  • hyperscale infrastructure deployments
  • AI-native infrastructure deployments
  • DCIM platforms
  • telemetry and monitoring systems
  • infrastructure-as-code tooling

What the JD emphasized

  • deep technical knowledge across the full stack (power, cooling, network) as it pertains to white space design and implementation
  • Demonstrated experience serving as an owner's engineer or resident engineer, including reviewing consultant drawings, managing contractor compliance, and interpreting construction specifications and submittal documents
  • Proven ability to manage multiple concurrent projects across different sites and work effectively with MEP engineers, general contractors, and equipment vendors in fast-moving environments.