builders, reviewers, and owners adopting AI without losing control
Practical AI
Agent safety, review workflows, automation guardrails, and field-ready AI use.
- Published
- 18 notes
- Latest
- Jun 27, 2026
AI evaluations need the harness
Cara gives AI buyers and reviewers a checklist for reading evaluation claims: disclose the harness, tools, budget, retries, validity checks, and failure rules.
A practical safety checklist for coding agents
Cara turns autonomous coding-agent safety into a reusable checklist for scope, tools, diffs, tests, provenance, and release gates.
AI browser agents need a safe browsing budget
Cara explains how to give AI browser agents useful work without letting tabs, credentials, downloads, forms, and external links become quiet production risk.