VEXXHOST reflects on ALL IN 2026 in Montréal, sharing conversations around AI infrastructure, GPUs, Kubernetes, data residency, and production readiness.
ALL IN 2026 brought together organizations, researchers, technology teams, and decision-makers from across the AI ecosystem for several days of conversations in Montréal.
For VEXXHOST, we were excited to connect with organizations exploring AI from different angles and at different stages of adoption. The conversations gave us a closer look at what teams are trying to build, where infrastructure is creating friction, and what matters most as AI moves from experimentation toward more serious production use.
From the pre-event cocktail on September 15 through two full days on the show floor, the week was a valuable opportunity to listen, exchange ideas, and share how the right infrastructure foundation can support what comes next.
Starting the conversations early
For some of our team, the week started on September 15 with the pre-event cocktail in Montréal.
It was a chance to reconnect with familiar faces, meet new people, and kick off what would become a packed few days of conversations around AI, infrastructure, research, and technology.
Photo credits: ALL IN
Those early conversations set the tone for the days ahead. Organizations are thinking seriously about what comes after the pilot stage and what it takes to support AI in a way that is reliable, scalable, and sustainable.
Two days of conversations on the show floor
On September 16 and 17, our team joined the wider ALL IN community for two full days of conversations, demos, and meetings.
We spoke with people from startups, enterprises, research organizations, the public sector, and technology teams working through a wide range of AI-related priorities.
Some conversations focused on GPU access and capacity. Others centered on Kubernetes, storage, networking, data location, operations, or the challenges of moving workloads into production.
That range was one of the most valuable parts of the event.
There was no single AI problem and no single infrastructure answer. Different organizations are moving at different speeds, with different requirements, constraints, and expectations.
For VEXXHOST, that is exactly where our experience across cloud infrastructure, OpenStack, Kubernetes, Ceph, networking, managed operations, and AI-ready environments becomes relevant.
What organizations are trying to solve
A clear theme across the event was that access to compute is only one part of the picture.
As AI environments grow, teams have to think about how workloads are scheduled, how resources are shared, how data moves, how performance is monitored, and how infrastructure changes are managed over time.
The questions quickly become broader:
- How do you scale without creating unnecessary operational complexity?
- How do you give teams access to the infrastructure they need while keeping visibility and control?
- How do you support new workloads without locking yourself into a platform that may not fit future requirements?
These are the kinds of questions we work with every day.
VEXXHOST’s AI infrastructure approach builds on open technologies and infrastructure expertise to help organizations support GPU-intensive workloads while maintaining flexibility across compute, storage, networking, Kubernetes, and managed operations.
Canadian infrastructure and control remain important
Another theme that stood out was the importance of infrastructure control.
For Canadian organizations, where workloads and data live can be an important part of the AI conversation. Data residency, infrastructure location, portability, and long-term control all become more relevant as AI moves closer to business-critical systems.
That aligns closely with how VEXXHOST approaches cloud and AI infrastructure.
As a Canadian-based and Canadian-owned provider, we can support organizations looking for Canadian infrastructure options while continuing to build on open-source technologies such as OpenStack, Kubernetes, and Ceph.
That open foundation matters because AI will continue to evolve.
Models will change. Accelerators will change. Deployment patterns will change.
Organizations need infrastructure that can evolve with them rather than becoming another constraint.
A booth that kept busy
Not every conversation started with infrastructure.
Our claw machine quickly became one of the more popular parts of the booth, giving attendees a chance to stop, try their luck, and have some fun between sessions and meetings.
It also turned out to be a great conversation starter. Some visitors came for the prizes and stayed to talk about what they were building, while others stopped by after seeing the activity from across the show floor.
Photo credits: VEXXHOST team
We also ran two Meta glasses giveaways each day, adding another highlight to the booth experience.
Congratulations again to the winners, and thank you to everyone who took part.
Photo credits: VEXXHOST team
Photo credits: VEXXHOST Team
Between demos, giveaways, technical questions, and informal conversations, there was always something happening.
Why being part of ALL IN mattered
As a sponsor, being part of ALL IN gave us the opportunity to contribute to the conversations shaping AI in Canada while staying close to the organizations working through real infrastructure decisions.
It also gave us a chance to showcase the areas where VEXXHOST can add value, from GPU-ready infrastructure and Kubernetes to open cloud platforms, storage, networking, data residency, and managed operations.
Just as importantly, it gave us the opportunity to listen.
The challenges organizations shared throughout the event reinforce something we see across the industry: successful AI adoption depends on more than models and accelerators.
It depends on the infrastructure underneath them.
Taking the conversations forward
We came away from ALL IN 2026 with new relationships, useful perspectives, and a clearer view of the infrastructure priorities organizations are working through.
The event also reinforced where VEXXHOST is well positioned to help: building open, flexible, and production-ready infrastructure for organizations that want to move AI initiatives forward without giving up control.
A big thank you to everyone who met with us during the cocktail evening, stopped by our booth, joined a conversation, watched a demo, tried the claw machine, or entered one of the giveaways.
And thank you to the ALL IN team for bringing so many parts of Canada’s AI ecosystem together in Montréal.
We’re leaving ALL IN 2026 with new conversations, stronger connections, and plenty of ideas to carry forward.
