A beautiful fall in New York City
I had the privilege of taking a break between my job at Cursor and my next role during the fall. Historically I have always gone straight from job to job, mostly to keep health insurance coverage (thanks America), and because I generally fear the unknown of not knowing what to do with my time without going to work every day. But this break taught me that even without work the world still goes on, and I can easily find things to do.
I took the time mostly to check off items from my personal TODO list that had remained on hold for the past year or longer. I bought a road bike, and had fun biking around New York City in the beautiful mild and dry fall that we had. I bought a 3d printer from Prusa, which I spent several days assembling and then many more days modeling and printing things to fix minor annoyances around the apartment. On several occasions I day tripped out to nearby towns, just to do small hikes or walks in nature.
I worked on side projects too, mostly Tasky, which absorbed many of my other efforts into one agent-friendly form-factor. I also revived an old idea of mine, where I can use a social network that isn’t public to organize the relationships in my life. This one was exciting to see how agents can read over my journal and populate the initial version, with dates and events in my life I enjoyed with those people. I also spent many tokens on a game, just for me, which I found more fun to build than to play so far.
What I’ll do differently this time
Before I jump back in to work, I want to record a couple issues for me in the workplace that I’m watching out for. Over the past year, I did a bad job of allocating time properly for my life outside of work. I shouldn’t need to wait for a break between jobs just to take care of myself and do the things I want to do.
I also will carefully watch how the constant juggling of agents affects my ability to focus. I believe that something has changed about my own ability to stay on tasks and drive them to completion, and am worried that the (now) years of AI coding are contributing to the problem. What I’ll start with first is to interact with fewer agents, which are possibly responsible for coordinating and doing more of the context switching on their own.
I’m reminded about some productivity advice I read long ago to just focus on doing the one most important thing. There might be many code changes in flight, but instead of switching back and forth between them, I’d love to just focus on the most important one, even if it means I have to wait, or avoid an additional dopamine hit from sending off another agent message while something completes. The intention of thinking harder about the current task, and making sure that I get to 100%, even if there are now five more that need my intention, I think will lead to better outcomes this time.