Thanks for stopping by our booth at IBM TechXchange.
Four days of conversations with platform, SRE and infrastructure teams running hybrid cloud. This page collects the themes, the demos and the next steps, so you can pick up where we left off or take it back to your team.
What we heard at the booth
The inner loop got faster. Operations didn't.
Everyone is building agents. Few can govern them.
Trust in AI SRE is earned in stages.
Shared staging is the bottleneck.
Getting the estate onto Terraform, and trusting the code.
Hybrid is the reality, and so is tool sprawl.
Six themes from the show floor
We took notes on our booth conversations. These themes came up again and again, from regional banks and insurers to retailers, telcos and systems integrators.
The inner loop got faster. Operations didn’t.
AI coding assistants are raising pull-request volume. Teams told us the queue now builds after the merge: review, release and incident response.
A team already building with IBM Bob said development is moving fast; operations is the half they still need to solve.
Everyone is building agents. Few can govern them.
Internal agent frameworks, AI governance reviews, internal marketplaces for agents. The question has moved from whether to use agents to how to give them context, scoped access and reuse across teams.
A head of platform at a large telco wants agents that are governed and reusable across the organization, not built in silos.
Trust in AI SRE is earned in stages.
Root-cause analysis first. Then fixes tested away from production. Then bounded action on low-blast-radius incidents. And per-agent permissions, never keys to every cluster.
A cloud platform leader at a global insurer described it as crawl, walk, run: prove the diagnosis before any fix goes near prod.
Shared staging is the bottleneck.
Developers named the shared staging environment as their biggest blocker, often before we asked. The follow-ups: dependent pull requests, CI outages and self-hosting for regulated teams.
A developer on a four-person team said every feature waits its turn for staging. An environment per pull request landed immediately.
Getting the estate onto Terraform, and trusting the code.
Unmanaged resources in Azure and GCP. AI-generated imports nobody could trust. App teams who won’t learn Terraform, and module changes that still need IAM and security review.
An infrastructure engineer had tried using AI to import existing resources, but had no way to know whether the generated code was safe to apply.
Hybrid is the reality, and so is tool sprawl.
On-prem, mainframe, IBM Power and Z run alongside public cloud. In the same breath: five or more observability tools, and bills that grow with every gigabyte ingested.
A platform engineer at a health insurer is working out what to do with on-prem Power systems while the rest moves to public cloud.
Schedule a session with your team and see how these themes play out in your own environment.
The Autonomous Operations Factory, product by product
Turn intent into spec-driven, reliable and auditable cloud. Here is what we demoed for each stage, and the question we were asked most.
Download the AOF whitepaper
Autonomous Operations Factory
One operating model across build, operate, observe and remediate, running on Aiden OS and governed by policy at every step.
Aiden for InfraOps
We showed bringing existing cloud resources under Terraform, a module change raised from a GitHub issue, and a security scan on an AI-generated Terraform pull request.
“Can it bring unmanaged resources under Terraform in a way we can review and trust?”
Shipyard by StackGen
Ephemeral environments for developers and their agents: a full environment for every pull request, so nothing waits in line for shared staging.
“What happens when two pull requests depend on each other?”
Aiden for DevOps
Pipeline-native delivery: agents that work inside your existing pipelines to make releases faster and safer as change volume grows.
“How do we keep review quality up when AI is writing more of the pull requests?”
Aiden for Observability
Managed open source observability with AI copilot capabilities, built to cut tool sprawl rather than add another dashboard.
“How does the pricing model compare with paying per gigabyte ingested?”
Aiden for SRE
Hypothesis-based root-cause analysis in Slack or Teams, the full causal chain, and fixes tested in an ephemeral environment and opened as a pull request, with per-agent access limits.
“How is this different from the SRE agents the cloud providers now offer?”
Pick up where we left off
Book your post-show demo
See what you missed, on your own environment. Every booked demo gets a 1-in-10 chance to win the LEGO Millennium Falcon.
Read the whitepapers
The same papers we handed out at the booth: the Autonomous Operations Factory and AI SRE.
Get the downloads
