Iro AIWhat you'll be able to do
- Clear your inbox faster with reusable email prompts
- Turn an hour of messy notes into clean summaries and action items
- Plan and prioritize your day with an AI thinking partner
- Build a personal library of prompts you actually reuse
- Know which tasks to hand to AI and which to keep
Inside the path
A focused set of five-minute lessons. Each one ends with a hands-on exercise, not a quiz you can guess.
Find your repeat tasks 4 min
Spot the handful of things you do every week that AI can take a first pass at.
Inbox in half the time 6 min
Prompts for triaging, drafting, and replying in your own tone, not robot-speak.
Notes to action items 5 min
Turn call notes, docs, and threads into a clear summary and a to-do list.
Plan your day with AI 5 min
Use AI to prioritize, time-block, and unstick decisions without overthinking.
Save it so it compounds 5 min
Build a reusable prompt library so today's win becomes next week's default.
Try a sample exercise
This is the kind of card you'd practice inside Iro: you do the thinking, then get feedback.
◆ Sample exercise · Prompt practice
It's Monday and you have 40 unread emails. You want AI to help you triage, but "summarize my emails" isn't something you can paste an inbox into. You need a prompt you can reuse on any single thread.
Your task: Choose the reusable prompt that will actually save you time on every email.
- "Summarize my inbox."
- "Here's an email thread. In 2 lines: what do they want, and what's the decision or reply needed? Then draft a short, friendly response in my voice that I can edit. If anything's unclear, list the questions instead of guessing."
- "Reply to this."
- "Make me more productive."
See why the second prompt wins
The winning prompt is reusable (you paste any thread into it) and it does the two things email triage needs: it extracts the ask and the decision, then drafts a reply you can edit, and it refuses to guess when something's unclear (which is where AI usually creates work instead of saving it). "Summarize my inbox" isn't actionable; this is a workflow you run 40 times. In Iro you'd build and refine prompts like this and get feedback on what makes them reusable.
Why most "AI productivity" advice doesn't stick
The internet is full of AI-tool listicles, and they mostly don't move the needle. The reason is that productivity isn't a tool problem; it's a workflow problem. Ten new apps you open once don't save time; three prompts you run every single day do.
The people who actually get hours back pick a few recurring tasks (email, notes, planning) and build a reliable prompt for each. Then they save those prompts and reuse them. That's the whole game: turn a good one-off result into a default you reach for without thinking.
What to hand to AI — and what to keep
- Hand off: first drafts, summaries, reformatting, brainstorming options, triaging what needs your attention.
- Keep: the final decision, anything with real stakes, your actual voice, and the judgment call on whether the output is right.
AI is a fast assistant, not a replacement for your thinking. The productivity comes from letting it do the 80% that's mechanical so you spend your energy on the 20% that isn't.
How do you turn an hour of meeting notes into a plan in five minutes?
Run the notes through three passes instead of one, and don't ask for a summary until the last pass. Here's the sequence, using a real hour of rambling notes.
1. Extract before you summarize
Paste the notes and ask: "List every decision made, every open question, and every commitment someone made, with who made it. Quote the line each one came from. Don't add anything that isn't in the text." The quote requirement is the step people skip, and it's the one that makes the whole output checkable in about ten seconds.
2. Separate what's yours
Follow with: "Of those commitments, which are mine? Group the rest by owner, and mark anything that has a date attached." You now have a task list instead of a wall of prose, sorted by who has to move next.
3. Write the update last
Ask for "a five-line update for people who weren't in the room, using only the decisions listed above." Summaries drafted first tend to smooth over the disagreement that was the entire point of the meeting.
Then check it. Skim the quotes, and if a line has no quote behind it, the model filled a gap on its own. Numbers and dates are where that happens most often, so those get a second look before anything reaches your calendar.
Which AI productivity habits quietly cost you time?
Four habits eat into the time AI was supposed to save, and none of them feel like mistakes while you're in them.
- The rewrite spiral. You ask for a draft, don't like it, ask again, and twenty minutes disappear. Give it two rounds. If the third is still wrong, you're missing context, not prompts: name the fix ("shorter, no adjectives, end on a question") or write the opening line yourself and tell it to match your voice.
- Generating what a template already solved. A weekly status update doesn't need fresh generation. It needs a saved format with three blanks in it. Starting from scratch every Friday is slower than filling in a form.
- The summary nobody checked. Summaries fail quietly. A dropped caveat leaves no hole behind it, so the text reads fine and you never notice what's missing. Length is the tell: when the output is a tenth of the source, something got decided for you.
- Typing context you could have pasted. Describing a document in your own words takes longer than pasting the document, and it loses detail on the way.
What ties these together is effort that isn't matched to stakes. A throwaway internal note and a client email don't deserve the same scrutiny, and treating them alike is what makes people give up on the habit. Verify hard where a mistake would be expensive, and skim everywhere else.