Topic Cluster
Prompting and craft
Prompting patterns, evaluation habits, and the craft underneath the systems.
4 articles in this cluster.
Part of the fuller People Ops AI: the guide for People and HR teams pillar →
Make one output reliable
Pick a task whose result someone can check. Give the model the source material, explain the required result and define what it must not infer. Save examples of good and bad outputs so changes can be tested rather than judged by how fluent they sound. This cluster is about the working method, not which workspace to buy.
Check the answer before widening the scope
Evaluate factual accuracy, omissions and usefulness on the same small set of examples. Change one part of the instructions at a time and record what improved. When a prompt starts needing persistent data, permissions or several handoffs, move from an isolated prompt to a workflow or system. Better wording alone cannot supply missing ownership.
Prompting patterns for People Ops
Better AI output is not about a better model. It is about prompting patterns for People Ops: five blocks, prompt chains, and the…
From prompts to systems
Buying tools or nudging people to use ChatGPT is not building AI capability. The move from prompts to systems has one order…
Choosing AI models for HR work
No single best AI model for HR work exists. Match the model to the task, run two or three, and cross-check what matters. Here is…
AI workspace setup for People teams (Claude, ChatGPT, Copilot)
Set up an AI workspace with standing instructions, workstream projects, approved references and review rules. The same pattern…
Reading is one thing. Doing is another.
The Grain Audit maps one People Ops process end to end, ranks the highest-return automations, and hands you a 90-day plan you keep whether or not we work together. Two weeks. £2,000, credited in full against a programme. Three slots a month.
Book a Grain AuditNot sure where your function stands yet?Take the Readiness Assessment→