Start Where You Are

By
•
Start Where You Are

Getting started with AI is easier than you think.

AI helps the most when you stop trying to use it for everything

If you took the class, you have already done the hard part. You know what AI is reasonably good at, what it is not, and how to keep yourself out of trouble with sensitive data. But what exactly is this thing going to do to help me?

You sit back down at your desk on Monday morning. The work has not changed. The inbox is the same inbox. And the question quietly becomes: okay, now what?

It's not easy to come up with use cases in the abstract. "How could I use AI?" is a hard question to answer in a vacuum. "How could I use AI on the three things I am actually working on this week?" is much easier, and it produces ideas you might actually try.

Paste the prompt below into the chatbot you already use. It asks you one question about your job, then comes back with seven to ten suggestions sorted by complexity — simple one-prompt tasks, slightly more involved sequences, and a few more advanced setups using Projects, Artifacts, or connected drives. Each one includes the actual prompts you can use as a starting point.

I want to find some practical ways to start using AI in my job. To give me
useful suggestions, please ask me one question first: what's my job, and what
does a typical day or week look like? I'll share my role, industry, and the
kinds of tasks that fill my time.

Once I answer, suggest 7–10 ways I could use AI in my work, ranging from very
simple (one prompt) to more advanced workflows. Focus on beginner-friendly use
cases like:

Simple (one prompt):
- Researching topics quickly (summarizing articles, explaining unfamiliar
  concepts, comparing options)
- Drafting documents in specific formats (emails, reports, meeting agendas,
  summaries, job descriptions)
- Rewriting or improving text I've already written (making it clearer,
  shorter, more professional, or adjusting tone)
- Brainstorming ideas, names, talking points, or approaches to a problem
- Explaining or summarizing complex documents, data, or jargon in plain
  language

Slightly more involved (2–3 prompts):
- Research, then draft — gather information on a topic, then use what AI
  found to write something (e.g., research a prospect, then draft an
  outreach email)
- Draft, then refine — create a first version, then ask AI to adjust tone,
  shorten it, or tailor it for a specific audience
- Brainstorm, then build out — generate options first, then pick one and
  develop it further (e.g., brainstorm meeting agenda topics, then draft
  talking points for each)
- Summarize, then act on — condense a long document or email thread, then
  draft a response or next steps based on it
- Analyze, then recommend — paste in information (notes, feedback, data),
  ask AI to find patterns or themes, then suggest what to do about them

More advanced (using Projects, Artifacts, or connected drives):
- Projects — set up a dedicated workspace for recurring work (e.g., a client
  account, a course I teach, a product I manage) where I upload reference
  materials, style guides, or background documents once and have AI use that
  context across every conversation
- Artifacts — have AI build something I can iterate on and reuse, like a
  reusable template (proposal, performance review, status report), a simple
  interactive tool (a checklist, calculator, or decision tree), or a
  polished document I can refine over multiple turns
- Connected drives (Google Drive, etc.) — let AI pull from my actual work
  files to summarize a folder of meeting notes, find information across
  multiple documents, draft something based on real source material, or
  compare versions of a document

For each suggestion:
- Describe the use case in one or two sentences
- Note whether it's a one-prompt task, a multi-step workflow, or uses a more
  advanced feature (and which one)
- Give me the example prompt(s) I could copy and paste to try it right away
  — if it's multi-step, show each prompt in order
- For advanced suggestions, briefly explain how to set it up (e.g., "create
  a Project, upload your style guide, then ask...")
- Keep the language plain — no jargon, no technical setup beyond what the
  feature itself requires

One reminder before you run it: the suggestions it gives you are starting points, not commitments. Pick the two or three that match something you are actually doing this week. Try those. Ignore the rest until they become relevant. The point is to get one or two real wins under your belt, not to build the perfect AI workflow on day one.

And of course — apply the green/yellow/red habit from the class to anything you paste into a prompt. Useful use cases on sensitive data still need the right tool and the right tier.

The people who get the most out of AI after a class are the ones who picked a small, real task on Monday morning, got a useful result, and then did it again on Tuesday. After a couple of weeks of that, the more advanced workflows stop looking intimidating, because you are no longer learning AI in the abstract. You are learning your AI, on your work.

Start where you are. The good stuff comes from there.

Ready to go deeper? Bring a teammate to the next AI Lunch and Learn session, or bring it to your team as a private cohort.

Back to Blog