If we all have the same number of hours in a week, why do some people seem able to squeeze so much more in?

Take Stephen King, a prolific writer known for churning out 2,000 words every single day, weekends and holidays not excepted. Or Marie Curie, who ran a lab and raised two daughters largely alone after her husband's death, and still became the only person to win Nobel Prizes in two different sciences.

Neither had extra hours, but they did have something even more powerful: a system. No one is that productive without one.

I’ve always been a big believer in using systems to make myself more efficient, whether it’s time boxing or batching similar tasks together so I'm not switching contexts ten times a day. As someone who values productivity, I’m finding all sorts of ways to use generative AI to get even more out of my days. Here are a three prompts I’m using constantly.

There are few things I despise more than a full calendar or stacked to-do list where nothing actually gets accomplished. To me, a day crammed with fake work on throwaway items is a day wasted.

That doesn’t mean I don’t understand why people do it, though. Humans crave control; when we tick off an easy-to-achieve to-do list item, we’re rewarded with a pleasant hit of dopamine. Unfortunately, nothing substantial has ever been accomplished through a series of meaningless tasks.

That’s why, as a founder who loves a good productivity hack, I was intrigued by recent chatter about how AI is revolutionizing the “backwards calendar,” a well-worn concept refreshed with an LLM-based twist.

The idea is simple: start with a goal you want to achieve, like writing a keynote speech or landing a specific client. Then, work backwards, breaking the project into tasks that can be accomplished in the days leading up to your final due date.

It’s a straightforward premise, and it works. The problem is it involves a lot of set-up—or at least it used to. Now, you can simply tell ChatGPT what you’re hoping to do, and ask it to act as a master scheduler to create a realistic timeline of concrete steps to get you there.

I tried this myself last week. My goal was closing out a product decision that had been dragging for two weeks. Working backward, the AI flagged something I'd missed: I'd blocked time to "think about it" three separate days, but never once scheduled the actual conversation with my product lead that would let me decide. This is where LLMs really shine: forcing you to clarify your goals and the steps you plan to take to accomplish them, with no room for excuses or fuzzy thinking.

Ah, 3 p.m. You’re five meetings deep in the day, your coffee buzz faded hours ago and now, and your post-lunch crash is hitting hard. Time for some deep work!

Hardly. It’s no secret that our energy levels rise and fall throughout the day, but what may surprise you is how much they matter for productivity. Researchers have found that trying to force a morning person into a late-night brainstorming session is a recipe for disaster—the same goes for night owls being expected to be at their best first thing. It’s what’s called a “chronotype mismatch,” and while it’s not something you can just push through.

Nor should you just be expected to focus all day, non stop. “We can’t expect to lift weights nonstop all day, and we can’t expect to use sustained focus and attention for extended periods of time, either,” Dr. Gloria Mark, a professor of informatics at the University of California, Irvine, told the New York Times.

To make the most of your peak times, ask ChatGPT for help. First, spend a few weeks documenting your energy peaks and valleys, making an hourly note of your focus and motivations on a scale of one to 10. Enter your findings into AI, and ask it to build your calendar around what it finds, moving deep work into your sharp windows and pushing meetings, email, and anything that doesn't need your best thinking into the dips.

Cal Newport wrote about “the shutdown ritual” back in 2009, and it's stuck with me ever since. At the end of every workday, he does the same thing: he checks his task list, notes what's unfinished, glances at tomorrow, then says a fixed phrase ("schedule shutdown, complete”) out loud to mark the day closed.

Lately, I’ve been doing my own version with ChatGPT. At the end of the day, I dump whatever is still open—a call I have yet to return, a decision I haven’t made—and ask it to sort these unfinished tasks into two piles: what needs to be done first-thing, and what can wait. What I get back is a single starting point for the next morning, so I am not opening my laptop trying to remember where I left off.

Before AI, I was confident that my systems were maximizing my efficiency, and making me the most productive version of myself possible. I was wrong. With just a few simple prompts, my calendar feels less bloated than ever, giving me back more time than I had thought possible. I’m still not writing 2,000 words a day like Stephen King, or winning any Nobel Prizes, like Marie Curie. But I do know that I am using my time wisely, and that’s good enough for me.