Dear Techies,
You already know which task it is
You don't need me to tell you.
There's something in your week, maybe your day, that you do over and over. It's not hard. It's not creative. It just takes time. And every time you do it, some part of your brain is thinking: there has to be a better way.
And there is a better way. But it's not the AI you've been using.
Chatbots wait. Agents work.
Most people use AI the same way. You ask a question. You get an answer. You ask again. You get another answer. The moment you stop talking to it, it stops.
That's a chatbot. It's reactive. It waits for you.
An AI agent is different. You give it a goal. It figures out the steps. It connects to the tools it needs, things like your email, your calendar, a spreadsheet, the web. Then it works through the task until it's done or until it needs your input.
It's the difference between asking someone a question and hiring someone to handle a project. One responds. The other acts.
Here's what's changed. Twelve months ago, building an agent required a developer. Today you can do it with plain English instructions on a platform with a free plan. No code. No technical background. If you can write an email, you can build an agent.
How to spot where agents fit into your work
Most people hear about agents and skip straight to the tools. They sign up for a platform, stare at a blank screen, and don't know where to start.
Here's a better starting point. A simple framework.
Spot it.
What tasks in your week are repetitive, predictable, and follow a pattern you could explain to someone else? Not the creative work. Not the judgment calls. The stuff that runs on an invisible checklist you've never written down.
Describe it.
Could you explain this task to a new colleague in under sixty seconds? If yes, you can describe it to an agent. If you'd struggle to explain it clearly, the task probably needs your judgment more than it needs automation. That's a useful filter.
Delegate it.
What would change if this task just happened without you? Correctly. Consistently. Every time. How much time comes back to you in savings? What would you do with it?
That framework is the starting point. Not a tool. Not a platform. Just three questions that tell you whether a task is worth handing to an agent or not.
What this actually looks like
Let me give you an example.
Every Friday you pull numbers from three different places, paste them into a document, write a short summary, and send it to the same group of people. It takes forty-five minutes. It's important, but it's not interesting. And you've been doing it the same way for months.
An agent could check those three sources, pull the numbers, draft the summary in your tone, and have it ready for your review by Thursday evening. So you just spend five minutes checking it instead of forty-five minutes building it.
That's not a fantasy. That's a real use case people are running right now with tools that have free plans.
Run it through the framework. Is it repetitive and predictable? Yes.
Could you explain it to a colleague in under sixty seconds? Easily.
Would it change your week if it just happened? Absolutely.
That's how you know it's an agent task.
"But what if it gets things wrong?"
Good question. And the honest answer is: sometimes it will.
Agents aren't magic. They follow the instructions you give them, and if those instructions are vague, the output will be too. Sometimes they'll misinterpret a step. Sometimes they'll do something you didn't anticipate.
That's why good agents have guardrails. You decide what they can and can't do. You decide when they need to check with you before acting. It's less like handing control to a machine and more like delegating to a very fast, very focused assistant. One who follows your rules exactly, but needs those rules to be clear.
So the risk isn't agents going rogue. It's vague instructions and an agent doing its best with them. Learning to write better instructions is the fix. That's a skill, not a talent, and one we'll build together as this series continues.
One thing to do before next week
Open a note on your phone or a blank document and write down three tasks from your week that pass the framework test.
For each one, write a single sentence describing what you'd want an agent to do.
Not how. Just what.
For example:
"Summarise my team's Slack messages from the last 24 hours every morning before I start work."
"Every time I get an invoice by email, save it to my accounting folder and log the amount in a spreadsheet."
"Before every client call, pull up the last three emails with that person and give me a one-paragraph summary."
Those three sentences are worth more than you think. When we get to the build stage later in this series, you'll already have your project brief written.
Want to see this in action?
This week's video walks through real examples of AI agents on screen. What they do, how they work, and what it actually looks like to set one up. The visuals make the concept click in a way that words on a page can't.
If you're curious, the video is the natural next step.
👉 Watch the video: https://youtu.be/W0ZpB4SFyic
Before you go
If this was useful, share it with someone who keeps saying they'll "get around to learning AI properly." This is the series that actually makes that happen.
Stay Savvy,
Ijeoma | Tech Savvy Starts Here
P.S. Did you know I wrote a book? Tech Savvy Starts Here is available on Amazon—a practical, engaging guide for families and educators helping kids build confidence with technology. Check it out here.
Enjoyed this edition?
Forward it to a friend or colleague who will enjoy it as well.
Missed something? Catch up in the newsletter archive.
🧠 Keep learning. | 💬 Keep questioning. | 💥 Keep growing.
