Author Image

Hi, I am Igor

Igor Widlinski

HPC Systems Engineer @ IREN | GPU Cloud · Bare-metal Kubernetes · Slurm · InfiniBand · NVIDIA/AMD Fleets

I build and operate the infrastructure behind large-scale AI compute — taking GPU data centers from bare metal to production: provisioning thousands of NVIDIA and AMD GPUs, automating the full server lifecycle, and running the networking, scheduling, Kubernetes, and storage platforms that AI customers run on. I focus on reproducible, infrastructure-as-code systems that keep large GPU fleets reliable at scale.

Certified Information Systems Security Professional (CISSP)
Certified Cloud Security Professional (CCSP)
Artificial Intelligence and Machine Learning
AWS Certified Solutions Architect – Professional
AWS Certified Solutions Architect – Associate

Experiences

1
IREN

Apr 2025 - Present

Canada

IREN is a next-generation data center company, building and operating 100% renewable-powered data centers for AI Cloud, GPU clusters, colocation, and HPC workloads.

System Engineer

Apr 2025 - Present

Responsibilities:
  • Automated bare-metal deployment of NVIDIA (B300/B200/H200/H100) and AMD (MI355X/MI350X) GPU fleets across multiple data-center sections (~1,000+ GPUs each) — OS imaging, configuration management, and out-of-band (BMC/Redfish) management.
  • Validated GPUs at scale (NCCL/RCCL testing, burn-in, firmware) and resolved hardware faults fleet-wide; delivered production GPU clusters to external AI customers.
  • Configured and troubleshot the InfiniBand fabric with NVIDIA UFM Enterprise — managed the subnet manager, topology/routing, and partition keys (pkeys) for tenant isolation; used UFM telemetry to diagnose congestion, link/port errors, and transceiver faults end to end.
  • Operated multi-petabyte storage arrays from VAST, Weka, and Ceph.
  • Owned the DCIM system of record — administered inventory-system access for the broader engineering org and led large-scale data-quality audits and remediation, reconciling inventory against live hardware across thousands of servers.
Wawanesa Insurance

June 2023 - Apr 2025

Canada

The Wawanesa Mutual Insurance Company, founded in 1896, is one of Canada’s largest mutual insurers, with over $4 billion in annual revenue and assets of 12 billion.

Senior Security Systems Administrator

June 2023 - Apr 2025

Responsibilities:
  • Deploying and maintaining servers in AWS with CloudFormation.
  • Developing strategies and plans to enforce security requirements and address identified risks.
  • Configuring security systems, analyzing security requirements, and recommending improvements.
  • Managing 5000+ endpoints using industry best anti-malware, DLP, and PAM services.
2
3
Sanctuary AI.

Jan 2023 - Jun 2023

Canada

Sanctuary AI is on a mission to create and deploy millions of industrial-grade humanoid robots to help address growing labor challenges.

Cloud System Administrator

Jan 2023 - Jun 2023

Responsibilities:
  • Implementing CI/CD capabilities for infrastructure and applications.
  • Contributing to the architecting and implementation of cloud-native and cloud migration solutions.
  • Writing infrastructure automation scripts and templates and integrating them with DevOps tools.
  • Infrastructure as Code using AWS CDK.
Zymeworks Inc.

March 2013 - January 2023

Canada

Developing novel medicines for patients with difficult-to-treat cancers and other serious diseases.

Senior Systems Administrator

March 2013 - January 2023

Responsibilities:
  • Migrated on premise datacenter to AWS.
  • Managed cloud ( AWS / Azure ) infrastructure.
  • Salt / Puppet / Terraform configuration management.
  • Created Gitlab CI/CD pipelines.
  • Zabbix monitoring and created Grafana dashboards.
  • Single Sign on using Azure AD / SAML / ADFS and various MFA solutions.
  • Created auto-scalable High Performance Computational Cluster (GridEngine) in AWS.
  • Worked with various storage technologies including NFS / AWS Lustre / Netapp.
  • ManagedMeraki / Palo Alto / Juniper networking.
4
5
Incomm Canada Prepaid Ltd.

October 2011 - March 2013

Canada

Incomm is the industry leading marketer, distributor and technology innovator of stored-value gift and prepaid products.

Linux Systems Administrator

October 2011 - March 2013

Responsibilities:
  • Created OS hardening scripts for PCI (Payments Card Industry) certification.
  • Server configuration, installation & maintenance (RHEL 5,6 and Solaris Patches and Firmware Updates).
  • Troubleshooting outages and coordinating resolutions with external entities.
  • Created OS update strategy for company’s older software/hardware.
Eigen Development Ltd.

July 2008 - October 2011

Canada

Successful transaction processing company dedicated to the task of managing payments for merchants.

Linux Systems / Network Administrator

July 2008 - October 2011

Responsibilities:
  • Server configuration, installation, & maintenance (Ubuntu).
  • Implementation/maintenance of VoIP server (Asterisk PBX) & all VoIP phones
  • Designed, supervised and implemented major power and environmental upgrade to the server room
  • Implementation of disaster recovery plan & daily back ups
  • Collaborate with developers for debugging and systems related issues
  • Performed network and application security testing as part of Payment Card Industry (PCI) certification.
6

Featured Posts

Hero Image
SAP-C02

Last Friday I passed SAP-C02 the AWS Solution Architect - Professional exam. This is my second AWS certificate having passed the AWS Solution Architect - Associate couple of months earlier. SAP-C02 took me about 2 months to study for and I used the following resources to study: Cantrill’s AWS Certified Solutions Architect - Professional course. This course was almost 70 hours long, and it took me about 1 month to go through. It is a good course, but gosh is it long… I feel like it would be much better to complete 3 courses that are ~20 hours long instead of doing single Cantrill’s class. Stephene Maarek’s AWS Certified Solutions Architect Professional 2025 course. I ended up doing this course twice, with the second time being 3 days before the exam ( this involved 5+ hours per day cram sessions). Highly recommend My own AWS Organization created just for learning and practice. This org is made out of 3 accounts, with SSO set up ( using Azure). Tutorial’s Dojo Tests. These tests are not perfect, and sometimes frustrating ( due to grammar, and the way they are written). For ~$15 USD the value is unbeatable. My test scores were as follows: Review mode Set 1: 73% Review mode Set 2: 76% Review mode Set 3: 76% Stephane Maarek’s Udemy SAP-C02 Practice Tests. My scores were: Practice mode Test 1: 72% Practice mode Test 2: 52% ( not sure what happened here, but I was sick and feverish when I did this one) Neal Davis’ Udemy SAP-C02 Practice Tests. My scores were: Test #1: 74% Test #2: 71% Test #3: 82% Overall the test took me 2 hours and 40 minutes to write, and overall was tough, but I felt confident I passed it when it was all done.

Tuesday, March 25, 2025 Read
Hero Image
First Post - initial environment setup

Plan Plan is to get AWS Certificates and at the same time: do all practicing and testing in code or cli (or as much as possible) update this blog with experiences/things learned get better at markdown, python, aws cli, etc Root Credentials AWS makes it clear that credentials for root account are not recommended. This makes sense, but we can not use aws cli without active credentials. Once we have the credentials, it will be possible to create new admin user and use that accounts credentials instead of root’s.

Saturday, January 4, 2025 Read

Recent Posts

Hero Image
SAP-C02

Last Friday I passed SAP-C02 the AWS Solution Architect - Professional exam. This is my second AWS certificate having passed the AWS Solution Architect - Associate couple of months earlier. SAP-C02 took me about 2 months to study for and I used the following resources to study: Cantrill’s AWS Certified Solutions Architect - Professional course. This course was almost 70 hours long, and it took me about 1 month to go through. It is a good course, but gosh is it long… I feel like it would be much better to complete 3 courses that are ~20 hours long instead of doing single Cantrill’s class. Stephene Maarek’s AWS Certified Solutions Architect Professional 2025 course. I ended up doing this course twice, with the second time being 3 days before the exam ( this involved 5+ hours per day cram sessions). Highly recommend My own AWS Organization created just for learning and practice. This org is made out of 3 accounts, with SSO set up ( using Azure). Tutorial’s Dojo Tests. These tests are not perfect, and sometimes frustrating ( due to grammar, and the way they are written). For ~$15 USD the value is unbeatable. My test scores were as follows: Review mode Set 1: 73% Review mode Set 2: 76% Review mode Set 3: 76% Stephane Maarek’s Udemy SAP-C02 Practice Tests. My scores were: Practice mode Test 1: 72% Practice mode Test 2: 52% ( not sure what happened here, but I was sick and feverish when I did this one) Neal Davis’ Udemy SAP-C02 Practice Tests. My scores were: Test #1: 74% Test #2: 71% Test #3: 82% Overall the test took me 2 hours and 40 minutes to write, and overall was tough, but I felt confident I passed it when it was all done.

Tuesday, March 25, 2025 Read
Hero Image
Oh My Zsh

Today I learned about Oh My ZSH. I’ve got it installed and running on my laptop. The reason to get it is because Oh My Zsh will not make you a 10x developer…but you may feel like one. Which hit the spot. Oh My Zsh makes Zsh sleek and I ended up changing my default SHELL to ZSH, but I do not plan to switch exclusively to ZSH as BASH is still the SHELL to be used on servers, and everywhere else.

Tuesday, March 25, 2025 Read
Hero Image
First Billing Alarms

Enabling IAM access to Billing According to this post giving IAM access to Billing Console can only be done through console. So let’s enable it.

Saturday, February 1, 2025 Read