← Home About Photos Replies Favorite Books Other Blogs Archive Links
  • AI as a Teammate

    I was reminded this week about a couple of quotes from Ethan Mollick’s blog regarding generative AI at work that match my personal experience:

    The first is from March 2025, explaining a randomized controlled trial testing office workers working with AI and without AI. Their conclusion:

    This suggests that AI effectively replicated the performance benefits of having a human teammate – one person with AI could match what previously required two-person collaboration.

    The second is from 2023 where Mollick discusses the results of an experiment he ran with professional consultants:

    Consultants using AI finished 12.2% more tasks on average, completed tasks 25.1% more quickly, and produced 40% higher quality results than those without.

    Nuance matters: what AI model you use (use a paid model), what tasks you use it with (text-based office work), and how experienced you are with the models (use it daily!). The conclusion is AI significantly expands, improves, and accelerates average office work. The trick back in 2023–and still today–is to learn how generative AI can improve your specific workflow. The only way to figure that out is using it and testing it yourself.

    → 7:58 AM, Sep 20
  • Leadership North Stars

    I did an introspective exercise recently that I found insightful for understanding how my brain approaches my work. I was told to come up with my “leadership north stars,” or statements that speak to the foundational assumptions I use when I’m leading others. And, me being me, I had trouble thinking of my north stars on my own, so I used chatGPT. Here’s the prompt I gave it:

    I’m doing an exercise to help me figure out my north stars of management. Those are the principles that I lead my team by. They serve as foundational ideas for my management style and provide me truths that i can rely on when I’m overwhelmed or stressed. Please ask me questions one at a time to help me deduce my north stars. Wait for my answer before asking another question. Please use my answers to influence what question you will ask next.  After you have enough information from me, tell me what you think my north stars are.

    I found the app’s questions to be quite helpful, overall. When I gave an uninspired answer to its question, it didn’t let me off the hook–it asked me further questions to drill deeper. I took its suggested north stars for myself and whittled it down to this list. If anyone wants to understand how I lead, this is a good starting point.

    My North Stars for My Leadership

    I trust in the intelligence and ability of every team member I hire. I trust in their competence and judgment, and I empower them to own their responsibilities.

    I stay engaged with my team’s work by asking thoughtful questions and keeping informed about their progress.

    I make myself available to support my team while encouraging them to solve problems independently.

    I willingly step into challenging roles when it benefits my team. I lead with a service-oriented mindset, allowing my team to direct my efforts to best support their success.

    I hold every team member to the same standard (assuming any needed reasonable accommodation)

    I value diverse perspectives and seek consensus by gathering opinions from all relevant voices, even if I ultimately trust my own judgment when making decisions.

    I commit to continuous personal growth, asking for specific feedback from others and seeking out best practices so that I can be the best leader for my team.

    I care for my staff as whole people by modeling a healthy work-life balance—ensuring that while I never drop the ball on important work tasks, I always prioritize my family and encourage them to do the same.

    → 4:13 PM, Mar 19
  • Productivity Trap with Oliver Burkeman - Big Think

    youtu.be/-LMaT4rUd…

    ⁠Perfectionism for me has always been a kind of central part of what I’m struggling with. That sense that I’m kind of on the back foot, that I need to put in just a little bit more, maybe a lot more effort and self-discipline, find the perfect organizational system, and, like, then I would finally get into the driving seat of my own life. It’s becoming very obvious that this ever-accelerating treadmill isn’t going to lead finally to this moment of wonderful calm and peace of mind. There will always be too much to do. You’re never going to feel completely ready. You’re never going to be able to feel confident about what’s coming in the future.

    It is hard to sum up a feeling of being one step behind achieving your best better than this quote from Oliver Burkeman. “There will always be too much to do” speaks to soul of my working life. And yes, I too am guilty of downloading every new productivity app!

    ⁠Attempt instead to start from the idea that you don’t need to do anything with your day to justify your existence on the planet. You may need to do it in order to earn your wage or your salary, absolutely, but you don’t need to do it for these existential reasons.

    This trap of feeling like your work gives full meaning to your life has unconsciously sapped joy from me. Because of my educational upbringing, nothing in this video is intellectual news to me. But intellect and action have been so very different in my life. The feeling of always being behind because the to-do list has grown longer instead of shorter; guilt over never finding the time to give your full attention. Let it go! Be at peace! Thanks for the reminder Oliver.

    And by the way, he has a pretty great blog and newsletter.

    → 10:05 AM, Feb 12
  • A good piece by Taylor Lorenz on the responses to the healthcare CEO shooting. Link www.usermag.co/p/yes-we-…

    → 7:37 AM, Dec 7
  • A short pedantic explainer on why Sam Altman’s note-taking advice is trash. Perfect holiday week content. www.theverge.com/2024/11/2…

    → 6:15 PM, Nov 25
  • You can now search the web in ChatGPT and it works pretty well on a first impression! Lots of legal and financial issues open ai will need to solve but it is really easy and quick for the user. I’ll be using this a lot!

    → 3:33 PM, Oct 31
  • I came across this old article in HBR about leaders being good at strategy or execution. They state, “only 16% of top leaders were rated very effective at either strategy or execution.” But look at the chart–it is also possible that people just don’t like their CEOs!

    → 3:43 PM, Oct 23
  • I just realized that my Reader app (by Readwise) can make quote images–cool! And I just found the new mission statement of my blog! Link

    → 11:58 AM, Oct 9
  • Here is a great write up from Elena Rossini on her reasons for being wary of the big social media platforms. A nice nugget:

    Social connections became gamified and quantified for the purposes of surveillance capitalism, mediated by Big Tech companies; platforms gave the illusion of offering a service that connects people but made their money from tracking them and serving them ads. People online are “users” (the same language used to describe drug addicts) and content appears in “feeds” (language employed to describe livestock).⁠⁠

    Link here

    → 3:00 PM, Oct 3
  • A great video on Chicago’s plan to end homelessness by building more affordable housing funded through a housing transfer tax. We already know affordable housing is crucial to ending homelessness, and funding it is a significant hurdle. I’m glad to see an original idea in this space though I’m sure there are pros/cons to consider.

    → 12:51 PM, Oct 3
  • Instagram is going to automatically dub videos in reels into your language while “maintaining the creator’s voice and even changing the movements of their mouth to match.”

    I’m excited for this to be everywhere–it is hard to NOT be excited to be able to communicate with anyone regardless of language. Translation is one of the best features of current AI and is something so obviously beneficial to all of us. From the verge

    → 5:13 PM, Sep 25
  • Raycast is probably the second best reason to buy a Mac over Windows for general knowledge-work computing (the first is probably the M1/2/3 chips). It basically gives anyone a simple way to do “power user” things on their mac just by typing in whatever they want–it makes my work life far easier and quicker. Raycast is going to windows and iOS devices. Windows makes sense, but it is not entirely obvious how Raycast will work on iPhones with the limitations Apple puts on apps, but I’m interested to see.

    → 8:25 AM, Sep 25
  • Ethan Mollick on the new model, GPT-o1, that OpenAI released yesterday:

    Using GPT-o1 means confronting a paradigm change in AI. Planning is a form of agency, where the AI arrives at conclusions about how to solve a problem on its own, without our help. You can see from the video above that the AI does so much thinking and heavy lifting, churning out complete results, that my role as a human partner feels diminished. It just does its thing and hands me an answer. Sure, I can sift through its pages of reasoning to spot mistakes, but I no longer feel as connected to the AI output, or that I am playing as large a role in shaping where the solution is going. This isn’t necessarily bad, but it is different. As these systems level up and inch towards true autonomous agents, we’re going to need to figure out how to stay in the loop - both to catch errors and to keep our fingers on the pulse of the problems we’re trying to crack. GPT-o1 is pulling back the curtain on AI capabilities we might not have seen coming, even with its current limitations. This leaves us with a crucial question: How do we evolve our collaboration with AI as it evolves? That is a problem that GPT-o1 can not yet solve.

    Every new turn of the LLM and AI story reinforces the idea that this is a huge change to work and intellectual life. What does it mean that a cheap model can do our logical reasoning for us and how can we use that to enhance our efforts? In the short term I think we can raise our expectations of everyone to now know the latest in research or best practices in their field. And if someone doesn’t, it means they aren’t spending the 5 minutes to query chatGPT or the like.

    → 8:34 AM, Sep 13
  • From Ken Saxon’s newsletter

    I forgot how challenging it can be to receive a big download of anonymous feedback, and to try to make constructive use of it. As I’ve done my whole life, I skipped right past all the nice things people said and focused on the critiques. When I do that, I find it’s easy to fall into defensiveness or make excuses.

    Fortunately, I had an executive coach who debriefed the assessment with me. She made sure I took in the complimentary feedback, and she helped me process the critical feedback in the constructive way it was intended — to support my development and our organization’s mission.

    I feel this same way each time I receive feedback! Feedback is powerful but is often be really tough to hear.

    → 6:00 PM, Sep 12
  • Getting Podcast Transcripts

    Why I want podcast transcriptions

    I often find myself listening to podcasts about homelessness, addiction, or other topics that relate to my work. A lot of times, I want to highlight parts of these podcasts to share in conversations or to chew on later to sharpen my thinking on the subject. To do that I prefer having the text available.

    Easy and costly options

    Just pay for Snipd or Matter

    • Snipd is a podcast app that works fine, but it’s not cheap. On iOS, you’re looking at $12 a month.
    • Matter is a read-it-later app that you can send your podcasts to. It’ll take a bit more effort than Snipd, but it’s easier on the wallet, starting at $8 a month.

    Just pay for Reader and hope the podcast is on YouTube

    Reader by Readwise is great for many reasons. It’s my go-to for read-it-later and general online-knowledge-management help. When it comes to podcast transcripts, they can pull in YouTube videos and grab the transcript. So if your podcast has a YouTube version, you’re golden - just import it to Reader and you’re set. Reader will set you back $9 a month on iOS. On Reddit, they’ve hinted that podcast transcription support is coming down the pipeline, so fingers crossed for that.

    Using Reader with YouTube podcasts is by far my preferred method.

    Kind of cheap, little scary, but useful!

    If a podcast isn’t on YouTube, here’s my backup plan:

    Get the transcript

    Option 1: Use an app like MacWhisper, Aiko, or Vibe to get a free transcription of the audio file. You’ll end up with a txt file, but heads up - it won’t be neatly formatted into paragraphs. The punctuation is often good but not perfect. If you’re cool with reading it as-is, you can stop here. If you want it more readable, we’ll need to rope in another AI to polish it up.

    Option 2: Use the Apple Podcast transcription. Just highlight all of it, right-click > services > open.

    Run the script to create an epub

    The transcript is accurate, but it’s not formatted well enough to read easily. So I run a Python script to clean it up. That sounded intimidating to me at first, but it’s actually pretty simple:

    1. Make sure Python is installed on your computer. If it’s not, go ahead and install it.
    2. I’ve got a script that Claude 3.5 Sonnet helped me create, you can find it on my GitHub. Download it and make sure to add your own Claude API key where it says [insert API key] at the beginning.
    3. In your terminal, navigate to where you put the script, drop the transcript .txt file there (name it something like podcast.txt), create an output folder, and then run the script. Make sure you’ve got the dependencies installed first. On my Mac, I run the script with this command, but specify the input file name and output directory name: python3 podcast-transcript.py [input_file] [output_directory]
    4. The script should do its thing and then ask you what title and author you want for the epub.
    5. You’ll end up with 2 usable outputs: combined_context.txt and whatever you named your epub. If you get stuck with any of this like I did, asking an AI is your best bet! I like getting an epub because it plays nice with Reader.
    → 2:00 PM, Sep 12
  • The best RSS and feed app, Reeder, just announced a new version! It seems like it’s an everything feed app, which is a course change for them. I’m excited to try it iOS App Store link

    → 10:41 AM, Sep 7
  • AI apps and API

    API is the way to go

    I’ve moved from paying OpenAI for their ChatGPT app and using the free version of Anthropic’s Claude to using both of their APIs. It takes more setup than downloading the ChatGPT app and paying your $20 a month. However, most people will probably use less than $5 a month in API fees. Most chat AI apps also allow you to input multiple API keys. Currently I have both Claude from Anthropic and OpenAI API accounts and use them both in apps on Mac and iOS. Its great to have two to compare between, while only spending $5-10 a month rather than $40 a month!

    Mac

    I use MindMac. You can use it for free for a while but then it is a one-time purchase. MindMac allows you to put in your API keys from multiple AI LLM services. It seems well thought out and gets regular updates. There are some free alternatives but they aren’t as easy to use. LibreChat was one I tried for a few minutes, but it requires you to install docker first.

    iOS

    I’ve been using Pal Chat. It is free app made by a redditor that allows you to use your API keys. It simple and easy to use, and easy to switch between AI providers. I’m using GPT-4o or Claude 3.5 Sonnet most of the time and its great on the app.

    iOS Shortcuts

    The APIs make ios shortcuts very flexible. For example, I made a shortcut that allows me to save a webpage as a “bookmark” in my notes app Logseq (could also easily do Obsidian). It gives me the options to send the webpage to openai to summarize the text and then pastes the summary in the note created in Logseq. One of the benefits of using the API rather than the ChatGPT app (besides cost) is it is so much faster in getting a response from OpenAi. Here’s the shortcut

    Things to note

    When using the API, you need to create your own intro prompt for the LLM. Most of the apps you pay for have a default one they use. What is nice about using your API keys and one of the apps above is you can customize your own prompt. For example, I have one I use when I have work-related questions and I have a general prompt. Here’s my general prompt:

    You’re a helpful assistant. Provide clear and thorough answers but be concise. If you don’t know the answer, just say you don’t know. You can ask me questions if that will help you give me a better answer. My career depends on you giving me a good and truthful answer.

    → 4:02 PM, Aug 31
  • We need a story about when times were hard and showed no sense of ever getting easier, and the heroic path was to believe against all evidence that they would. A story about the fact that there is nothing trite about everyday happiness, and that such happiness persists even when everything else declines. That the Dawnless Days will be followed by a sunrise if we just keep going, keep going, keep going.

    Discussing Tolkien in 2021, in light of the pandemic. Makes me want to read the books again! Link

    → 7:55 AM, Aug 31
  • Just testing using micro.publish with Obsidian.

    → 9:01 PM, Aug 28
  • Leadership and letting people leave well

    Quote from Nilay Patel of The Verge

    Happy to share this video and tell y’all to subscribe to Becca’s channel, I hope she kills it. The thing I am proudest of is that The Verge consistently hires people with no experience but obvious drive and ambition and gives them the space and freedom (and yes, irritating guidelines) that help them go be their own leaders, and then we celebrate their work no matter where they are. Becca is going to be one of the best examples of that yet.

    I think this post from Nilay is so profound—someone who gets that taking care of employees is not at odds with managing your organization. I would add that it takes courage to celebrate staff leaving for better opportunities since it is not the norm. I lay is probably a great leader and a self-aware manager to have this kind of insight. Link to the post on threads

    → 5:05 PM, Aug 28
  • Figuring out how to use micro.blog

    Things to Try out

    1. Cross-post to Mastodon
    2. Cross-post to Threads
    3. Follow Threads and Mastodon from micro.blog
    4. Post to micro.blog from obsidian
    5. Post to micro.blog from phone

    Obsidian and micro.blog

    • There’s a community plugin, micro.publish, that posts to micro.blog from obsidian link
    • Obsidian front-matter works with it!
      • title – the page or post title
      • tags – the category on micro.blog
    → 3:51 PM, Aug 25
  • RSS
  • JSON Feed
  • Micro.blog