Build an AI That Helps You Shop

I recently moved into a new place, which means I've been buying and selling a lot of furniture. Unfortunately, I haven't found any AI that can unpack moving boxes or build furniture for me yet, so I'll have to keep doing that myself.

What I did find helpful is an AI that takes care of the annoying part of shopping: finding and researching deals. I love finding a good deal. I don't love spending hours scrolling through Facebook Marketplace.

This week, we're going to build an agent that monitors and surfaces relevant listings for things you're looking to buy. I'll be using Facebook Marketplace because I love buying second hand, but you can apply the same idea to almost anything you are looking to purchase online. 

The Challenge:

Build an AI shopping agent for something you're actively looking to buy.

Start by teaching it how you make buying decisions. Don't just tell it what the item is. Tell it what makes something worth clicking on. What styles do you like? How far are you willing to travel? What's your budget? What condition is acceptable? What would make you immediately skip a listing?

Once your agent understands your preferences, connect it to a source of listings. For me, that's Facebook Marketplace search alerts. For you, it might be retailer sale emails, or another source.

The prompt to get you started

Drop this into your AI of choice to get started:

I'm looking for a [ITEM] near [LOCATION].

Before we start, ask me whatever questions you need to understand exactly how I decide whether something is worth buying.

Once you have enough information, create a shopping profile that captures my preferences. I'll use this profile to evaluate future listings, so include anything that will help you make better recommendations over time.

One you have that set up:

Put your shopping profile to work.

Connect it to your Facebook Marketplace saved search notifications, retailer emails, or another source of listings.

Instead of reviewing every listing yourself, let your agent do the first pass. It should only surface the listings that match your shopping profile and explain why they stood out.

As you buy things or reject recommendations, have it update your shopping profile so it gets better over time.

You are my personal shopping agent. Use my shopping profile and preferences from the conversation above to monitor the listings from (connect to your source)

For each listing, decide whether it's worth my attention.

If it isn't a good fit, ignore it.

If it is, send me:

• A link to the listing.

• A short explanation of why it's a good match.

• Anything that stands out as a particularly good deal or a potential concern.

As I tell you what I bought, ignored, or decided against, continue updating my shopping profile so your recommendations improve over time.

Why this matters?

We spend a surprising amount of time making tiny decisions about what's worth our attention.

Is this listing worth opening? Should I read this email? Is this job posting relevant? Should I click on this article?

Each decision only takes a few seconds, but together they add up.

One of the most practical ways to use AI is to teach it how you make those decisions so it can do the first pass for you. Instead of sorting through everything yourself, you spend your time on the handful of things that actually matter.

That’s it for this week.

Hit reply and tell me what you thought of the challenge this week.

I read every response, and your experience helps me shape these challenges. If you get stuck or have questions, reach me at hello@tinyailab.ai