Skip to content
FluidstackFluidstackSan Francisco, CA

Data Center Infrastructure Architect

Designs scalable optical and structured cabling for hyperscale data centers, mapping logical networks to physical fiber plants, planning transceivers, and ensuring lifecycle scalability. Requires expertise in MPO ecosystems, optics standards, and DCIM integration.

200k – 300k/yr
HybridDevOps / SRE

About the role

Responsibilities

Logical / Physical Mapping

  • Break down DCN logical port maps (TOR, spine, super-spine, DCI) into physical cross-connect and patching schedules.
  • Automate port-to-fiber mapping at scale (100K+ ports per site).

Fiber Plant Design

  • Define cassette types (MTP/MPO Base-8/12/16), trunk counts, and structured cabling hierarchy.
  • Plan cassette placement within racks, rows, and meet-me-rooms.
  • Model fiber utilization and overbuild for 2–3 hardware generations.

Optics / Transceiver Planning

  • Select optics (400G/800G DR4, FR4, LR4, SR8, coherent DCI) based on reach and loss budget.
  • Validate interoperability across vendors and generations.
  • Maintain loss budget and margin models for every link type.

Structured Cabling & Physical Implementation

  • Define patching architectures (direct MPO vs. breakout vs. mid-span).
  • Ensure pathways (ladder racks, underfloor, ODFs) scale with fiber density.
  • Establish standards for labeling, documentation, and change management.

Scalability & Lifecycle Management

  • Build modular designs for pod/rack growth.
  • Anticipate transition to higher-speed optics (e.g., 400G → 800G → 1.6T).
  • Integrate with DCIM (NetBox, custom tools) to track strands, ports, and patching.

Requirements

  • Proven experience designing hyperscale or cloud provider data center networks (1000+ racks, 100K+ fiber strands).
  • Expert in structured cabling systems (Corning, Panduit, CommScope) and MPO/MTP ecosystems.
  • Deep knowledge of optical standards (IEEE 802.3, OIF) and transceiver technologies (400G/800G DR4, FR4, LR4, SR8).
  • Strong grasp of fiber loss budgets, dispersion, and testing methodologies (OTDR, insertion loss, polarity verification).
  • Experience integrating designs into DCIM tools (NetBox, Remedy, custom hyperscaler equivalents).

Compensation

Base Salary Range: $200,000 - $300,000 per year, depending on experience, skills, qualifications, and location. Total compensation may include equity in the form of stock options.

Skills

Mtp/MpoStructured Cabling400G Optics800G OpticsDr4Fr4Lr4Sr8NetboxDcimOtdrIeee 802.3OifCorningPanduit Commscope

Similar roles

DevOps / SRE jobs
Fluidstack

Development Engineer, Facility Software Automation

FluidstackAustin, TX +3

Own and operate the containerized infrastructure, CI/CD pipelines, and observability stack for Fluidstack's critical telemetry and automation platform that powers gigawatt-scale AI compute facilities. Requires production ownership of Kubernetes, IaC, and reliability-focused observability.

200k – 250k/yr
On-site5+ YOEDevOps / SRE
Fluidstack

Platform Engineer

FluidstackSan Francisco, CA +3

Build foundational internal platforms (CMDB, DCIM, asset management, monitoring, automation) that power Fluidstack's global AI compute infrastructure and data center operations. Requires 3+ years building production systems with Python/Go, databases, IaC, and infrastructure/DevOps experience.

200k – 250k/yr
On-site3+ YOEDevOps / SRE
Console

Infrastructure Engineer

ConsoleSan Francisco, CA

Infrastructure Engineer building and operating scalable cloud and self-hosted platforms on AWS and Kubernetes. Own core systems for graph, search, and runtime while ensuring enterprise-grade reliability, security, and deployment for autonomous AI agents.

200k – 350k/yr
On-site5+ YOEDevOps / SRE
Together AI

Platform Engineer, Model Shaping

Together AISan Francisco, CA

Build and operate backend services and infrastructure for model customization and evaluation at Together AI. Requires 3+ years building production infrastructure, strong Python/Go skills, and deep experience with Kubernetes, Linux, and cloud platforms.

200k – 290k/yr
Hybrid3+ YOEDevOps / SRE
Kepler

Platform Engineer

KeplerNew York, NY

Own AWS infrastructure, Pulumi IaC, deployment pipelines, and security baseline for an AI research platform serving financial institutions. First dedicated platform hire defining enterprise deployment, SOC 2 controls, and developer experience.

200k – 280k/yr
On-site5+ YOEDevOps / SRE