Job Overview:
The Cambridge HPC Sysadmin is a field-based consultant that assembles end-to-end research computing system solutions. This role is designed for an ambitious, experienced sysadmin who wants to grow and embrace challenges not afforded by managing a single environment. They will leverage their expertise in scientific computing and knowledge of the technology landscape to drive outcomes that exceed client expectations.
Responsibilities and Duties:
Gather client requirements, design optimized solutions, sometimes using a single vendor's portfolio and more often using a broad variety of vendors and technologies.
Demonstrate knowledge of higher education, federal labs, research institutes, pharmaceutical companies, and bioinformatics organizations.
Leverage Cambridge’s brand as NVIDIA’s partner of the year for three consecutive years in higher education to meet the unique needs of their scientists and researchers.
Be intimately involved in the design, configuration, rollout, and troubleshooting of these environments as well as training staff and providing staff augmentation. This includes deployment of a new solution or augmenting an existing HPC/AI solution from the ground up.
Consult on and assist with day-to-day management of clients’ research compute infrastructure environments.
Maintain HPC/AI infrastructure in Linux-based environments for new and existing clients.
Lead technical discussions and be the face to the client in preparation for and during engagements.
Validate solution designs, meet client requirements, and are technically feasible and deployable.
Ensure solutions are simple and easy to understand while taking into account the client’s overall capabilities / skills.
Scope out and detail professional services deliverables setting clear client expectations.
Build documentation and provide knowledge transfer required for clients to support their environments.
Display expertise in storage, networking, data protection, digital archiving, and other infrastructure technologies.
Gain advanced expertise of and certifications from the vendors Cambridge uses in our solution stack.
Qualifications:
Candidates must have experience providing deployment services or cluster administration.
University undergraduate degree in Computer Science, Computer Engineering, or science related field required.
Because every environment is unique, candidates must have broad exposure to related technologies. This includes knowledge of GPU-focused hardware/ software and Linux system administration (package management, IP networking, troubleshooting etc.). They must also have knowledge of cluster design / management technologies (Bright, Werewolf, XCat etc.), storage technologies and parallel filesystems (Lustre, GPFS, BeeGFS etc.), networking and configuring network switches (ethernet and InfiniBand), familiarity with HPC schedulers (SLURM, UGE, LSF, etc.) and programming / libraries (MPI, CUDA, etc.), and understanding of Scripting (Bash, Python, etc.).
Have some knowledge of tech industry leaders including AMD, DDN, Dell, HPE, IBM, Intel, Juniper, Lenovo, Microsoft, NVIDIA, Oracle, Vast, VMWare, WEKA, and others.
While much of our work can be done remotely, in order to get exposure to this variety of environments, the roles involve about 50% travel, usually in short, few day trips in the US..
Candidates must have impeccable communication skills, an ability to multitask, and high attention to detail. They must be effective problem solvers, organized, creative, intellectually curious, deal with ambiguity, and able to work with different types of personalities.
Authorization to work in the United States on a full-time basis required.