Sovereign AI starts with infrastructure. Train, tune, and deploy on infrastructure you control. Own your AI stack.
United States
- Tweets 2,255
- Following 272
- Followers 994
- Likes 1,838
Starting in one hour: see how ELM controls which RPMs each host group can run.
Brian Rieb and Patrick Swartz cover sources, content views, version pinning, and handing installs and rollbacks to Ascender.
11 AM PT / 2 PM ET
Register: ow.ly/tapO50ZSrvn
19
CIQ is heading to #AiEverything Abu Dhabi, October 6 and 7.
Meet our team to talk about portable, sovereign AI infrastructure across on-premises and cloud environments.
📍 ADNEC Centre
Details: ow.ly/lEvS50ZSLmj
23
Who decides which RPMs your fleet is allowed to run?
Tomorrow, Brian Rieb and Patrick Swartz walk through Enterprise Linux Manager (ELM): sources, content views, pinning, and the hand-off to Ascender.
11 AM PT / 2 PM ET
Register: ow.ly/yXgl50ZSnKa
37
Open weights let you run a model. They do not tell you what trained it.
On Oct. 1, Gregory Kurtzer and Addison Snell explore why AI sovereignty requires verifiable training data and licenses, and how OpenWALDO is building that foundation.
Register: ow.ly/VtHZ50ZREr5
3
3
112
Heading to #EDUCAUSE2026 next week?
Connect with Rose Stein and David Horn in Denver to compare notes on secure, scalable infrastructure for research computing, AI, and other critical campus workloads.
Sept. 29 to Oct. 2
Details: events.educause.edu/annual-c…
2
88
We’re still energized by #TACCSTER2026.
CIQ was proud to sponsor the student talks. Congratulations to the student winners of the Young Researcher and Best Poster awards.
Thank you to the Texas Advanced Computing Center (TACC), which runs Rocky Linux on Stampede3 and Lonestar6, and the Rocky Linux community.
tacc.utexas.edu/taccster/202…
2
127
President Bjorn Hovland spoke with Hankyung in Seoul about why Korea sits alongside Taiwan and the US as one of CIQ's three essential markets for AI infrastructure.
Rocky Linux adoption keeps growing worldwide, and in Korea, enterprises like LG Uplus already trust it in production.
Read the full interview (in Korean) below:
ow.ly/7Z9p50ZR5JY
#RockyLinux #EnterpriseLinux #AIInfrastructure #OpenSource #CIQ #Korea
3
112
You'd never push code to prod without staging it first. Your Linux fleet does that every day.
On September 29, Patrick Swartz shows how to mirror, test, and promote changes to your whole fleet from one place with CIQ Enterprise Linux Manager (ELM).
Register (live or for the recording): ow.ly/wzHt50ZR2vp
3
94
Run Ollama on GPU infrastructure you choose, with your model, runtime, and data under your control.
See how RLC Pro AI, Ollama, and a neo cloud turn sovereign inference into a pull and deploy workflow with an OpenAI compatible endpoint.
Read more: ow.ly/OrYo50ZQK4A
#SovereignAI
1
99
Running RHEL, Oracle Linux, SUSE, Amazon Linux 2, Ubuntu, or another community distribution? There is a documented path to RLC Pro.
See what carries over, what you gain, and which migration path fits your environment: ow.ly/pMU050ZQEp4
#EnterpriseLinux #RLCPro
2
79
“Here’s WALDO: AI’s first open source project and community.”
On October 1, Gregory Kurtzer and Addison Snell will discuss open weights, AI Bills of Materials, enterprise sovereignty research, and OpenWALDO’s shared training corpus.
Register: ow.ly/pAJa50ZPjvN
5
131
From HIV research to helping scientists use HPC at NIH, Wolfgang Resch has followed the questions that led him to computing. Now he brings that experience to Fuzzball at CIQ.
Meet Wolfgang and hear his perspective on Fuzzball 4.2 and Arcee AI: ow.ly/g1Yv50ZPeIi
3
100
That’s a wrap on #AIInfraSummit! We shared what’s new with RLC Pro AI, Fuzzball 4.2, and our work with Arcee AI.
Congrats to DGX Spark winner Jesse Marshall of Recor Medical, whose idea uses AI agents to explore new uses for existing drugs and medical devices.
See you at #SC26!
1
5
204
The patch meant to fix a CVE can break production.
CIQ Enterprise Linux Manager (ELM) helps teams catch dependency conflicts in a draft, test the signed content view on a few hosts, and return to a prior version if needed.
Read: ow.ly/NEmM50ZOPcu
2
119
One hour to go.
An AI agent that sets up and runs real work on your cluster, under permissions you hold. That's Fuzzball 4.2's MCP agent loop, plus workflows that launch their own workflows and nodes that report their own health.
11 AM PT / 2 PM ET: ow.ly/8Nub50ZOw09
3
62
Your GPU budget deserves a kernel check.
Model loading, inference scheduling, and storage all depend on the OS beneath your AI stack. Brian Dawson explains what to look for and how RLC Pro AI addresses it.
Read: ow.ly/4vsA50ZOxbO
3
82
“Arcee has built the technical systems required to develop and operate models at significant scale. Fuzzball’s role is to give that team a consistent orchestration layer across different compute environments without requiring its training and inference stack to be rewritten for each provider,” said Bjorn Hovland, president of CIQ.
ow.ly/ZBWK50ZOors
5
102
“We have built our own training and inference stack because control over the full model lifecycle is fundamental to how Arcee operates. Fuzzball gives us an additional orchestration layer for coordinating workloads across heterogeneous compute environments without replacing the systems our team has built,” said Lucas Atkins, chief technology officer of Arcee AI.
ow.ly/9PlV50ZOo2C
4
121