erKam

Use when measuring how well an AI agent handles a recurring kind of task. Builds a small eval set, runs it, and raises the difficulty once the agent passes.

Use when letting AI agents run code, install packages or touch real systems. Covers sandboxing, least privilege and quarantine basics.

Use when structuring a codebase that AI agents will edit heavily. Weighs duplication against abstraction now that repeating code and keeping copies in sync is cheap.

Use when you have several independent jobs for AI agents. Splits work into parallel tasks, hands them off like to coworkers, and reviews results in batches.

Use when reviewing code an AI agent wrote, especially in a language or area you don't know well. Judges the result through tests and behaviour instead of reading every line.

Use when making an app or library usable by a person's own AI agent. Designs a command-line interface so agents can drive the app without a built-in chatbot.

Use when deciding whether an app should be native, web, or hybrid (Ionic/Capacitor) now that agents make native code cheaper to produce.

Use when giving an AI coding agent a job. Turns a task into a problem-and-outcome brief with acceptance checks, instead of step-by-step instructions.

Use when you want an AI agent to improve performance, size or startup time of a project over a long unattended run, with tests as a safety gate.