Operations | Monitoring | ITSM | DevOps | Cloud

On this episode of Masters of Data, we dig into what happens when AI actually delivers on its promise to eliminate busywork, and explore why removing the toil doesn't feel like the win everyone expected. We make the case that repetitive tasks build the intuition, pattern recognition, and muscle memory people need to do the harder work well. Security and engineering leaders rethinking how much triage and busywork to hand off to AI will find plenty to chew on here, especially anyone staring down a task list where every single item feels like the hardest one.

1,133 requests failed on one bad commit. The patch and the regression test are already written by the time anyone is asked to read the exact diff. This demo runs AURA as one step inside a Temporal workflow, alongside Codex. A GET request against a product catalog service goes from success to HTTP 500, and ClickHouse records the version, commit, trace ID, and exact error for every request. By the time AURA investigates, all 1,133 requests on that version have failed.

Open up an AI model and what's actually inside is a floating array of decimal points. No one can look at that and confirm it hasn't been tampered with, doesn't contain bias, or wasn't trained on poisoned data. This video covers why that changes how you need to think about trusting a model: If you can't unpick the model itself, you have to be able to trust its origin.

Checkout starts failing because the cart service keeps dying. The real cause is a recent pull request that let memory grow with every add to cart.

Give a model permission to act on your computer and you're trusting it will behave the way you expect. It might not. In this clip from our Braintrust conversation, Adam Berman, engineering leader at Semgrep, breaks down why a backdoored model is a different threat model than backdoored software. You can't fuzz-test your way to finding it, and you often can't detect it until it's already acting the way it shouldn't.

Matt Zytaruk, VP of Technology at Minga, on what changed once his developers could point their AI tools at their own infrastructure. Minga checks in 1.5 million students across the eastern seaboard every school morning. They run that platform on Control Plane, which gives their team an MCP server, a CLI, and Terraform-native resources — so a developer can investigate an issue and generate the code for the change without waiting on anyone.

Average throughput across all projects on CircleCI increased 59% YoY.

Observability looks perfect in a slide deck – in practice, it's messier. In this panel, engineering leaders from Cyara, PlayHQ, and National Australia Bank share what really happened when they scaled observability: unexpected cloud bills, alert fatigue, a weekend database outage caught by an AI-assisted triage agent, and a vendor dispute settled by a single chart. They also cover moving beyond legacy tooling, using AI to close the PromQL skills gap, and what's next – from agentic SDLC integration to continuous profiling. Real stories, real numbers, real lessons.

Password resets are a perfect starting point for agentic automation. See how Resolve Agent Lab builds an AI agent that verifies users, resets credentials, and closes the loop automatically.

Password resets are a perfect starting point for agentic automation. See how Resolve Agent Lab builds an AI agent that verifies users, resets credentials, and closes the loop automatically.