Getting started with AI: a beginner's guide to using Claude effectively

Many people new to AI stop at "ask one question, get one answer." But if you're building products, managing projects, or juggling many tasks a day like me, AI can become a genuine assistant — if you know how to use it right.

1. Give AI context, don't just issue a command

Instead of asking:

"Write a rejection email for a candidate"

Try:

"Write a rejection email for a candidate who made it to the second interview for a Product Owner role. Good skills but lacks fintech experience. Polite tone, leave the door open for future collaboration."

The answer will match your actual need far more closely, because the AI isn't guessing.

2. Hand off tasks that can be broken down, not a vague whole project

AI works best on tasks with clear boundaries: drafting, summarizing documents, comparing options, checking logic. For big tasks, break them into steps — the same way you'd split a backlog into user stories.

3. Always double-check, especially numbers and proper names

AI can write very fluently but slip up on details (dates, figures, company names). Before sending anything out, read it back like a tough reviewer.

4. Use AI to push back on yourself

A use case few people think of: hand the AI your plan and ask directly, "What are the holes in this plan?" or "If you were the toughest stakeholder in the room, what would you ask?" This is the fastest way to find weak spots before the real presentation.

5. Save the prompts that actually work

If a particular way of asking gets you a great result, don't let it slip away — save it as a template to reuse. Over time you'll build your own personal "prompt toolkit" for your work.

---

AI doesn't replace your thinking — it amplifies your execution speed. The best AI users aren't the ones who ask the smartest-sounding questions, but the ones who know most clearly what they want before they ask.