Stocked

The Year of Build: Project One

Shawn Farner

Shawn Farner

— 5 min read
The Year of Build: Project One

In every role I take on, I try to find ways to multiply my efforts. I want to get more done in less time. Then I spend the time I steal back seeking out additional ways to be productive. (I don't like being bored.)

Typically, I do this by building tools — web apps or scripts that reduce friction or speed up processes. LLMs have only made it easier to build such things. I can go from concept to creation in a fraction of the time.

Look... I have serious misgivings about "AI" in the creative realm. It can only dream what has already been dreamed. It lacks the soul and the spark to make anything new. It can't think outside the box. It is, by its nature, very much in the box. Until we're at such an advanced stage that "AI" truly thinks, feels, imagines — this will always be a limitation and the best it can do is copy and pretend.

But for deploying a tiny little app that saves me 20 minutes each time I use it? For that, it's geniunely great.

As I said, I usually focus these tools on work. Something I've not done to this point, though, is point the spotlight at my own life.

What could I do better? Where are there points of friction? What tasks could I speed up? What do I really dislike doing that I could offload onto a specialized tool?

The more I thought about it, the more I realized that there are loads of opportunities to apply this same thinking to my own personal day to day. So I decided to make a whole year out of it. There are 52 weeks in a year. My aim is to knock out 26 projects.

The first is...

Stocked: Caring about groceries and cooking so I don't have to

If we're going by elementary school report card grades, I get a solid "needs improvement" in anything food related.

  • I view grocery shopping as a chore.
  • I don't really plan or organize anything related to groceries or meals.
  • I don't track what I have well, which means I don't use items intelligently.
  • I order out A LOT.
  • I could definitely make healthier choices.

Stocked, in its current form, will hopefully help with all of these!

Here's how it works:

We start with some settings. I set a target daily caloric intake, a max spend per "trip," the ease (how long I want to spend preparing these meals), and the "Food Vibes" — types of food I like.
Next, I generate the recipes. This shoots out to ChatGPT (4o-mini if you must know) and comes back as properly strutured menu items with calories, required items and prep time. I select how many of each I want. Then: "Generate List."
A grocery list is generated, which looks at the recipes I choose and what's required for them. All of it is combined in a pretty smart way (not by me; I'm not complimenting myself).
I can then assign the recipe items I chose to the appropriate slots in any 15-day period. I don't have to think about it. All the choices are there already. I just plug them in.
When I grocery shop, I know what I'm getting and what it's for. When I put everything away, I log it. I can scan the UPC or enter manually. The UPC scan utilizes the iPhone camera and ties into the free API provided by OpenFoodFacts. This has a neat trick: there are "Unused" and "Open" sections and... it calculates the estimated expiration date for either state.

You may have noticed that all-too-familiar "AI" icon sitting there mysteriously in the corner. Right now... it actually doesn't do anything. ChatGPT figures in pretty heavily throughout the app anyway, but I am thinking about having "Smart Recs" appear at some point based on my Inventory, my eating habits and how I use things. Think "you're going through Coke Zero too fast" or "here are some ways you can use those eggs that are about to go bad." Those kinds of things.

I'm not married to the idea of including those Smart Recs, however. Ultimately, I'm treading carefully. I want anything that goes in here to be practical. If it's doing it just to do it — "because it's cool" — that isn't a good enough reason.

If this were a commercial product, there are things I'd give more time and attention to. I'd take several additional passes at naming menus and settings. I'd further scrutinize the layout. There'd be a smoother onboarding and a guided experience to take the user through each step.

But the truth is: I have zero plans or desire to distribute Stocked. There are likely apps that do this kind of thing already, but that's not even the reason I have. I honestly just don't want the responsibility of upkeep, which I'd take on the second another person who isn't me downloads it. And I'm really not interested in a side hustle right now. I just want to do this one thing better while freeing time up for the others I enjoy a lot more.

In the meantime, this first "Year of Build" project is basically done (just in time!), so now we're on to the next one. I know what it'll be. I'll give you the hint that it's somewhat related. You'll have to wait about two weeks to find out exactly what it is, though. See you then.

Read more