Senior GPU Cluster Deployment Technician – Hands‑On Infrastructure & Automation Specialist at brighthush
About brighthush
brighthush is at the forefront of the next generation of AI infrastructure. We deliver high‑performance, bare‑metal GPU clusters to AI and ML customers across a global, carrier‑neutral network of data centers. Our mission is to power the most demanding inference workloads with unmatched performance, price, and operational credibility. By leveraging our own Ethernet and InfiniBand fabrics, control plane, and data‑center‑scale automation, we can deploy a fully‑functional cluster in as little as 90 days—no greenfield campus required. As we expand to more than a hundred locations, we need a new breed of technicians who can turn raw hardware into production‑ready AI platforms with speed, precision, and a relentless focus on quality.
Why This Role Matters
Every cluster that brighthush delivers is a promise to our customers: a reliable, high‑throughput platform that meets or exceeds their performance expectations. As a Customer Deployment Technician, you are the final gatekeeper before a cluster enters production. Your expertise ensures that every node, every link, and every firmware version is validated, documented, and ready for the customer’s first day of use. In a hyper‑growth environment, your work directly influences our ability to meet delivery dates, maintain customer satisfaction, and scale our operations worldwide.
Key Responsibilities
- Hardware Validation & Burn‑In – Independently inspect every rack build, perform burn‑in tests, and verify firmware, BIOS, and driver baselines across all nodes.
- Network Verification – Confirm switch configurations, cabling, and link health on both InfiniBand and Ethernet fabrics; troubleshoot optics and fiber faults.
- Automation Execution – Run brighthush’s Ansible playbooks, interpret dashboard alerts, and resolve any automation failures that arise.
- Issue Diagnosis & Escalation – Identify root causes of hardware or network failures, document evidence, and coordinate with remote hands or on‑site crews to implement fixes.
- Performance Benchmarking – Execute cluster‑wide benchmarks (e.g., NCCL, HPL) and ensure results meet or exceed baseline expectations before handoff.
- Documentation & Runbooks – Maintain accurate NetBox records, log faulty parts for RMA, and author up‑to‑date runbooks for each build step.
- Continuous Improvement – Collaborate with Engineering to refine automation scripts, add new checks, and streamline the deployment workflow.
- Stakeholder Communication – Keep Project Managers informed of progress, raise any failures the same day, and hand over secure, documented clusters to Cloud Ops and Customer Success.
What You’ll Achieve in Your First 90 Days
- Days 1‑30 – Master our build process, validate your first racks, and run burn‑in and node tests on a live build.
- Days 31‑60 – Lead a full build from hardware validation to handoff, manage a second build in parallel, and deliver your first cluster to a customer.
- Days 61‑90 – Handle multiple builds simultaneously, take ownership of an additional build step, and formalize the process in written documentation.
Essential Qualifications
- 3–5 years of experience in Linux systems administration, server deployment, or data‑center operations.
- Proficient with optics and fiber troubleshooting, including low‑light and no‑link fault detection.
- Solid understanding of Ethernet networking and switch configuration; ability to trace faults from server port to switch.
- Comfortable with command‑line tools, scripting, and interpreting Ansible playbooks.
- Experience directing remote hands or field technicians to resolve hardware issues.
- Strong troubleshooting methodology: evidence‑based, step‑by‑step, and capable of explaining exclusions.
- Excellent written communication for runbooks and test records that others can follow.
Preferred (Bonus) Skills
- Hands‑on experience with GPU servers and the NVIDIA software stack; willingness to learn brighthush’s GPU test tools.
- Knowledge of InfiniBand or RoCE networking.
- Experience with BMC and Redfish interfaces.
- Familiarity with cluster benchmarks such as NCCL or HPL.
- Experience with Dell, Supermicro, or Lenovo servers and their BMC tools (iDRAC, IPMI).
Tools & Technologies You’ll Use
- Linux (Ubuntu, CentOS, etc.)
- Ansible playbooks and automation dashboards
- Redfish and BMC tools (iDRAC, IPMI)
- NetBox for asset and configuration management
- Zabbix for monitoring and alerting
- NVIDIA DCGM, NCCL, HPL for GPU performance testing
- InfiniBand and Ethernet fabrics for high‑speed networking
Career Growth & Learning Opportunities
At brighthush, your role is not static. As you master the deployment pipeline, you’ll:
- Take on more complex builds across diverse hardware vendors and network topologies.
- Lead the design of new automation modules, influencing how future clusters are provisioned.
- Collaborate with Engineering, Cloud Ops, and Customer Success to shape the end‑to‑end customer experience.
- Develop deep expertise in GPU technologies, high‑performance networking, and large‑scale data‑center operations.
- Earn recognition as a subject‑matter expert, opening doors to senior technical or leadership tracks.
Company Culture & Work Environment
brighthush thrives on speed, precision, and continuous improvement. Our distributed team operates in a flexible, remote‑first environment with occasional site visits (up to 25% travel). Build and test weeks may extend beyond normal business hours, and we value the ability to adapt your schedule to meet critical milestones. We empower our technicians to own their work, use AI tools to accelerate tasks, and always double‑check results before delivering to customers.
Compensation & Benefits
We offer a competitive base salary ranging from $144,000 to $192,000, commensurate with experience and skill level. In addition, you’ll receive equity participation, allowing you to share in brighthush’s growth. Our benefits package includes:
- Health, dental, and vision insurance
- 401(k) with company match
- Generous paid time off and flexible vacation policy
- Remote work flexibility with occasional travel support
- Professional development stipend and access to industry conferences
- Wellness programs and mental‑health resources
How to Apply
If you’re ready to bring your technical expertise to a fast‑paced, high‑impact role, we’d love to hear from you. Join brighthush and help shape the future of AI infrastructure—one cluster at a time.