Uploaded July 2026 | Updated September 2026, 1 day ago
In this AMA livestream, Viktor and Scott tackle a wide range of questions spanning DevOps careers, AI tools, infrastructure tooling, and platform engineering. They debate the right use cases for Helm with Crossplane compositions, weigh in on Backstage's Tech Insights plugin, and share strong opinions on when LLMs should — and shouldn't — replace deterministic scripting. A key theme throughout is using the right tool for the job, drawing a memorable parallel between the early Kubernetes hype cycle and today's AI-everything mentality.
The duo also offers candid career advice for those looking to break into DevOps, emphasizing the value of building fundamentals from the ground up — think Linux sysadmin and help desk roles — before jumping into cloud-native tooling like Kubernetes, Argo CD, and Terraform. They discuss AI infrastructure roles, the economics of SaaS in the age of agents, GitOps patterns for Terraform with tools like Atlantis, and the challenges of learning distributed inference at home. Whether you're a seasoned engineer or just starting out, this session is packed with honest, practical insights.
▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬
00:00 Intro (skip to first question)
08:12 Crossplane compositions with Helm: who is right?
11:38 Do you use Tech Insights in Backstage?
12:46 Should we pivot to helping LLMs instead of building tools?
17:13 Helm as OCI artifact for delivering compositions?
17:31 Should SPIFFE/SPIRE be part of Kubernetes product roadmap?
25:59 Should AI agents handle Terraform builds via Atlantis?
29:45 Career advice: 5 years DevOps experience without Kubernetes?
33:51 Can you get a DevOps job with no experience?
35:51 Advice for transitioning to DevOps from English major?
40:10 How to gradually grow into a DevOps career?
44:53 What do you think of AI infrastructure roles?
48:49 Is SaaS dead for indie hackers and solopreneurs?
50:26 Model plane (Kube-plane?) production adoption update?
50:55 How to learn AI infrastructure without expensive hardware?
52:15 How to split up a huge Crossplane composition?
55:10 Is it easy to get a DevOps job abroad vs other roles?
57:13 Should I use Terraform to install an AKS cluster?
58:26 How to handle legacy Terraform stacks with GitOps pattern?
01:01:55 How to find remote work and mentorship opportunities?
In this AMA livestream, Viktor and Scott tackle a wide range of questions spanning DevOps careers, AI tools, infrastructure tooling, and platform engineering. They debate the right use cases for Helm with Crossplane compositions, weigh in on Backstage's Tech Insights plugin, and share strong opinions on when LLMs should — and shouldn't — replace deterministic scripting. A key theme throughout is using the right tool for the job, drawing a memorable parallel between the early Kubernetes hype cycle and today's AI-everything mentality.
The duo also offers candid career advice for those looking to break into DevOps, emphasizing the value of building fundamentals from the ground up — think Linux sysadmin and help desk roles — before jumping into cloud-native tooling like Kubernetes, Argo CD, and Terraform. They discuss AI infrastructure roles, the economics of SaaS in the age of agents, GitOps patterns for Terraform with tools like Atlantis, and the challenges of learning distributed inference at home. Whether you're a seasoned engineer or just starting out, this session is packed with honest, practical insights.
▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬
00:00 Intro (skip to first question)
08:12 Crossplane compositions with Helm: who is right?
11:38 Do you use Tech Insights in Backstage?
12:46 Should we pivot to helping LLMs instead of building tools?
17:13 Helm as OCI artifact for delivering compositions?
17:31 Should SPIFFE/SPIRE be part of Kubernetes product roadmap?
25:59 Should AI agents handle Terraform builds via Atlantis?
29:45 Career advice: 5 years DevOps experience without Kubernetes?
33:51 Can you get a DevOps job with no experience?
35:51 Advice for transitioning to DevOps from English major?
40:10 How to gradually grow into a DevOps career?
44:53 What do you think of AI infrastructure roles?
48:49 Is SaaS dead for indie hackers and solopreneurs?
50:26 Model plane (Kube-plane?) production adoption update?
50:55 How to learn AI infrastructure without expensive hardware?
52:15 How to split up a huge Crossplane composition?
55:10 Is it easy to get a DevOps job abroad vs other roles?
57:13 Should I use Terraform to install an AKS cluster?
58:26 How to handle legacy Terraform stacks with GitOps pattern?
01:01:55 How to find remote work and mentorship opportunities?
