May 7, 2026

I Asked Claude. Then I Tried It. Here's What Actually Worked.

I use Claude pretty much every day. Not because I'm an AI person. Because I'm a tired person with too many open browser tabs in my brain.

Most of the AI content out there is written by people trying to sell you a course or convince you to start a business. This is not that. This is just the real list of things I use it for, what worked, what didn't.

What works.

Decoding school emails. The school sends a six-paragraph email with three different links and one buried sentence about a permission slip due Friday. I paste it in and say "what do I actually have to do." It tells me. I do it. Five minutes saved, fifteen minutes of resentment avoided.

Meal planning around what I already have. I open my fridge, take three pictures, and ask for five dinners using what's in there plus whatever I'd pick up at the store for under twenty bucks. It's faster than scrolling Pinterest and the recipes are actually doable on a Tuesday.

Drafting the hard message. The email to the coach about playing time. The text to my ex about a schedule change. The note to the teacher about a sensitive thing. I tell Claude what I want to say and the tone I want to land, and it gives me a draft that is one revision away from sendable. Saves me from sending the angry first draft.

Holiday and birthday gift ideas with constraints. "Eleven year old boy, loves baseball, already has a bat and a glove, budget eighty bucks, needs to be something he doesn't already have." Real list, fast, no Amazon rabbit hole.

Trip planning. Flights, hotels, ballpark addresses, weather, packing lists. All of it in one conversation. I keep the chat open and add to it as the trip gets closer.

Calendar wrangling. This is the big one. I tell it what I want on the calendar, in plain English, and it figures out the events, the times, the colors, the recurrence. It's like having a very fast, very literal assistant who never asks me to repeat myself.

What doesn't work.

Anything where I don't actually know what I want. If I'm using AI to avoid thinking, the output is bad. If I'm using AI to execute on something I've already decided, the output is great. The difference is whether I did the hard part.

Emotional processing. It's a fine sounding board for clarity. It is not a friend. I do not vent to it about my life. I vent to actual humans. The AI is for the logistics, the humans are for the feelings.

Recipes when I want them to actually taste good. It will give me a recipe. It will not give me a recipe that has been tested twelve times by an actual cook. For the dinners I care about, I still use real cookbooks and real food blogs.

Anything where being slightly wrong is a big deal. Medical, legal, financial. I use it to organize my thinking. I do not use it to make the decision.

The actual upgrade.

The thing that changed my life wasn't any single use case. It was realizing that the friction in my day is mostly tiny tasks that take fifteen minutes each because of decision fatigue, not because they're hard.

What's for dinner. How do I word this email. What should I get him for his birthday. What's the schedule for next weekend. These aren't hard problems. They're decision-fatigue problems. And AI is genuinely, weirdly good at removing decision fatigue.

If you're going to try one thing this week, try this: the next time you stare at your fridge and feel that low-grade dread about dinner, take a picture and ask. See what comes back. That's the whole pitch.

#AI-for-moms#systems#honest-mom-life