Latest / How I Work / Quick Win: The AI workflow that can replace your morning scroll
Transcript
- 0:01Speaker 1: What if every morning, instead of drowning in a sea
- 0:04Speaker 1: of newsletters, tweets, and podcasts, you had the day's biggest
- 0:09Speaker 1: breakthroughs already sorted for you. That's what Morgan Brown, vice
- 0:14Speaker 1: president of Product and Growth at drop Box, built for himself.
- 0:18Speaker 1: By the time he sits down with his coffee, an
- 0:20Speaker 1: AI workflow has already scanned the landscape of research, papers, substacks, podcasts,
- 0:27Speaker 1: and even YouTube. It spits out just three can't miss signals,
- 0:32Speaker 1: ranked from world changing to incremental, and explains what they
- 0:35Speaker 1: actually mean for his role. In today's quick Win, Morgan
- 0:40Speaker 1: shares the simple steps he used to design that system,
- 0:44Speaker 1: starting with one prompt, refining and running it daily. By
- 0:48Speaker 1: the end, you'll know how to set up your own
- 0:51Speaker 1: version to cut through the noise and actually stay ahead.
- 1:00Speaker 1: Welcome to How I Work, a show about habits, rituals,
- 1:04Speaker 1: and strategies for optimizing your day. I'm your host, doctor
- 1:08Speaker 1: Amantha Imber. Still staying on the topic of I guess
- 1:15Speaker 1: reducing or eliminating some of the grunt work through AI,
- 1:19Speaker 1: I'd love to know if they're more on the automation
- 1:23Speaker 1: or agentic AI side of things. What some of the
- 1:28Speaker 1: I guess other tasks that you've been able to automate
- 1:31Speaker 1: in interesting ways would be.
- 1:32Speaker 2: I've really done a lot of automation around trying to
- 1:35Speaker 2: understand what's changing around me. So I have a daily
- 1:39Speaker 2: workflow that reviews all new kind of substack newsletters, important
- 1:44Speaker 2: tweets across a large domain of AI and mL ex accounts,
- 1:49Speaker 2: reviews Archive, which is the scientific paper repository for new
- 1:55Speaker 2: mL and artificial intelligence papers. It scans the whole landscape
- 2:00Speaker 2: to Spotify podcasts and basically takes all of the day's
- 2:04Speaker 2: information about AI and mL reviews. It all categorizes it
- 2:09Speaker 2: into world changing, interesting, incremental, and noise, and then summarizes
- 2:15Speaker 2: the top three quote unquote world changing potentials and then
- 2:19Speaker 2: summarizes it for me in the context of my role
- 2:21Speaker 2: at DASH, where it kind of talks about hey, things
- 2:24Speaker 2: you might consider as the head of product for Dash. So,
- 2:27Speaker 2: for example, Anthropic the other day just announced that claud
- 2:30Speaker 2: can handle now a million token context window, which effectively
- 2:35Speaker 2: means that for many reasonable size software codebases, it can
- 2:39Speaker 2: reason about the entire codebase, which is a pretty big
- 2:41Speaker 2: and important milestone for software development. And the other one
- 2:44Speaker 2: is I automate my personal email I can't imagine you're
- 2:47Speaker 2: inbox Atmantha, but it gets hammered all the time. And
- 2:50Speaker 2: so I built a little agent using claud code to
- 2:55Speaker 2: read my personal Gmail every day, categorize it into newsletters,
- 3:01Speaker 2: spam priority emails, summarize the newsletters for me, flag the
- 3:05Speaker 2: messages that I want to respond to, and draft responses
- 3:10Speaker 2: for those newsletters or not for those emails. So really
- 3:14Speaker 2: trying to kind of like manage my inbound email automatically
- 3:18Speaker 2: as well.
- 3:19Speaker 1: Wow, okay, I love the both those examples. I want
- 3:21Speaker 1: to dig a bit deeper into how you went about
- 3:24Speaker 1: creating this automation that summarizes all the AI news of
- 3:30Speaker 1: the day. I mean, that sounds so powerful. Can I
- 3:34Speaker 1: buy that from you? Morgan? Are you selling it?
- 3:37Speaker 2: You can link it in the show notes?
- 3:39Speaker 1: Wow? Amazing handed over? Yeah wow. Talk me through the
- 3:43Speaker 1: process of how you put that together.
- 3:45Speaker 2: First of all, I realized that there were a few
- 3:47Speaker 2: key sources of information that I was constantly referring to personally,
- 3:51Speaker 2: So you know, whenever I had a free moment, I
- 3:53Speaker 2: would go to archive and kind of browse the actual
- 3:57Speaker 2: published papers because that's the real source of truth cutting edge.
- 4:01Speaker 2: I would see a lot of popular posts on x
- 4:05Speaker 2: about kind of new research breakthroughs. I follow some of
- 4:08Speaker 2: the key people like the sam Altman's of the World
- 4:11Speaker 2: and the Frontier Labs. And then there's also several great
- 4:15Speaker 2: substack newsletters around AIS that I read as much as
- 4:19Speaker 2: I can. But I just realized there was no way
- 4:21Speaker 2: for me to stay on top of that volume consistently,
- 4:24Speaker 2: and I just started with a prompt. I told chat TBT,
- 4:27Speaker 2: I'd like to create a prompt to scan the entire
- 4:30Speaker 2: AIML space for important signals around research and development and
- 4:35Speaker 2: industry news that pertains to my role at Dropbox. And
- 4:39Speaker 2: I said before X, I want to look across these
- 4:42Speaker 2: accounts and accounts like it. So I kind of gave
- 4:45Speaker 2: it a seed list of accounts, but asked it to
- 4:47Speaker 2: expand where it could. I gave it specific newsletters that
- 4:51Speaker 2: I read on substack, but also again asked to consider adjacents,
- 4:55Speaker 2: and then I gave it specific categories on archive where
- 4:59Speaker 2: those papers are housed and asked it to consider relevant
- 5:02Speaker 2: ones that there. And then I said, I'd like to
- 5:05Speaker 2: really filter high signal, like I don't want this to
- 5:08Speaker 2: be a laundry list of everything that happened that when
- 5:11Speaker 2: actually helped me kind of consume it any better. So
- 5:14Speaker 2: I said, help me come up with a scale to
- 5:16Speaker 2: rank the innovations and the news that's coming out in
- 5:20Speaker 2: the day, and kind of categorize things as you can't
- 5:22Speaker 2: miss these are like foundational important, shifts incremental and important,
- 5:27Speaker 2: and then give me some synthesis around Hey, when you
- 5:29Speaker 2: put these signals together, what does it mean for my role?
- 5:33Speaker 2: So I fed that all just in basically talking to
- 5:37Speaker 2: chat GPT. I like to use the audio input to
- 5:40Speaker 2: kind of just stream of consciousness my thoughts into it.
- 5:43Speaker 2: Once I did that, I asked it to kind of
- 5:45Speaker 2: structure it into a prompt. I reviewed the prompt. Some
- 5:48Speaker 2: of the things I didn't like. I always try to
- 5:50Speaker 2: go back and forth with it. I think you should
- 5:53Speaker 2: kind of refine it much like you would the copy
- 5:55Speaker 2: for a website or you know, the script for something.
- 5:58Speaker 2: And then once I got the prompt too you get start.
- 6:01Speaker 2: I said, Okay, let's run that for today and give
- 6:04Speaker 2: me the output. And I got the output and I
- 6:05Speaker 2: was like, oh, this is a little too long. It
- 6:07Speaker 2: could be a little shorter. It looks like we're underrepresenting,
- 6:10Speaker 2: we're missing YouTube. Let's try to add YouTube in, so
- 6:13Speaker 2: on and so forth, and then I finally got it
- 6:15Speaker 2: dialed in, and then I said, great, set this as
- 6:17Speaker 2: a daily task every day five am Pacific. Run it
- 6:20Speaker 2: again while I'm having coffee. I just kind of read
- 6:23Speaker 2: through it.
- 6:24Speaker 1: What I love about Morgan's approach is that it didn't
- 6:27Speaker 1: start with coding or fancy tools. It started with a
- 6:30Speaker 1: simple question, what sources do I keep checking and how
- 6:34Speaker 1: could I pull them together? So the next time you
- 6:36Speaker 1: feel buried under articles or updates, try listing your key sources,
- 6:40Speaker 1: feed them into a prompt and ask AI to filter
- 6:43Speaker 1: and rank what matters most. And of course you'll find
- 6:47Speaker 1: the full episode with Morgan linked in the show notes.
- 6:51Speaker 1: If you like today's show, make sure you git follow
- 6:54Speaker 1: on your podcast app to be alerted when new episodes drop.
- 6:58Speaker 1: How I Work was recorded on the traditional land of
- 7:01Speaker 1: the Warringery people, part of the Coulan nation.