Frigga Blogs

Claude can reason about a production incident. The problem is that production cannot answer back. The evidence is scattered across cloud resources, metrics, logs, traces, deployments, and source code. Debugging with AI quickly becomes a copy-past…

A repository is no longer where an AI coding tool's context has to end. Modern software rarely maps cleanly to one repo. A frontend may depend on backend services, shared libraries, workers, infrastructure code, and configuration spread across s…

A frontend calls an API. That API invokes another service. A queue triggers a worker somewhere else. One change can cross four repositories before the feature is complete. The problem isn't getting another coding agent. It's giving the agent…

Give an AI more code, and eventually the context itself becomes the problem. Across 10,000 files and millions of functions, there will always be something that looks relevant - another similarly named function, an outdated reference, a README menti…

Chaos engineering has a premise I find genuinely convincing. Tests only check the cases you thought of. Injecting real failure into a real system finds the ones you did not. It also has a problem sitting right at the front of it, and I have not seen i…

Every alert you have is built around something going wrong. A pod crashes, a request errors, a metric crosses a line. Something occurs, and the occurrence is what fires the alert. Now think about a nightly backup that just stops running. There'…

A reviewer opens a pull request and sees what changed. Files, lines, additions in green. What they cannot see is what happens next. Which services call this. Whether a mobile client parses that field strictly. Whether an overnight batch job reads the …