**Active Reliability Clearance Required**

Location: Ottawa, ON (on-site)
Language: English

About the role

We’re looking for an experienced System Administrator to assess, repair and maintain a group of high-performance, GPU-based Linux workstations and servers. A data science team uses these machines for proof-of-concept artificial intelligence (AI), machine learning (ML) and data science projects. Some of the systems currently have operational problems.

Your job is to find out what’s wrong, fix it, and leave the workplace secure enough for the team’s current and future AI and data science work. You’ll also check whether the current storage is enough for backups and upgrades, and recommend more if it isn’t.

What you’ll do

- Diagnose and fix system issues. Find the root cause of hardware, operating system, GPU and software environment problems, and fix them. This includes hardware failures, configuration problems, driver issues and slow performance on GPU workloads. If something can’t be fully fixed, recommend a repair, replacement or reconfiguration.
- Set up and test backups. Put proper backups in place and confirm they can actually be restored. Point out gaps or risks in the current backup and recovery approach.
- Maintain and upgrade Linux. Fix package errors, kernel issues, missing dependencies, failed services and stability problems. Apply security patches, upgrade operating system components, repair broken installations and tune configurations so they work with GPU computing.
- Fix the GPU and AI/ML software stack.



Resolve problems with NVIDIA drivers, CUDA Toolkit compatibility, GPU library dependencies, and frameworks such as PyTorch and TensorFlow, including container environments.
- Confirm the fixes work. Run GPU diagnostics, CUDA checks, test workloads and resource utilization checks to show that GPUs are detected, libraries work and the systems stay stable.
- Clean up dependencies and environments. Fix package conflicts, missing libraries, incompatible drivers and application configuration problems so the environment is stable and can be rebuilt the same way.
- Document your work. Record the issues you found, root causes, fixes, configuration changes, test results, anything left unresolved, and your recommendations.

What we’re looking for

- Strong hands-on Linux system administration experience: installing, monitoring, upgrading and maintaining operating systems, hardware and software
- Experience with NVIDIA GPU drivers, CUDA and GPU-accelerated computing environments
- Familiarity with AI/ML frameworks (PyTorch, TensorFlow) and container environments
- Experience with backup and recovery procedures
- Strong troubleshooting and problem-solving skills
- Ability to work with business analysts, project managers, developers and stakeholders to improve system performance
- Clear written documentation skills
- Fluent written and spoken English
- A valid Reliability security clearance

Pay: $650.00-$675.00 per day

Application question(s):

- Do you have active security clearance(Reliability, Secret)?

Work Location: In person

📌 System Administrator (Linux/GPU Systems) (Ottawa)
🏢 Makwa Resourcing
📍 Ottawa