River

Begin with one concrete task that matches River's main job: AI voice copilot for live sales conversations, product demos, follow-up, and revenue workflows. Keep the first test narrow and provide only the files, account connections, permissions, or context required for that task.

Run several realistic examples, including one edge case. Inspect the generated result or completed action for accuracy, missing context, unsupported assumptions, and recoverability. If the product works across other apps or company systems, verify what it can read and change before enabling broader access.

Keep human review for consequential customer, health, financial, security, production, or publishing actions. Where logs, traces, citations, approval steps, or evaluation tools are available, use them to understand why the AI produced an answer or took an action.

After results are reliable, save reusable instructions, templates, policies, or workflows and expand access gradually. Re-test important workflows after major model or product changes.