Maybe this is a case of holding a hammer, and everything’s a nail.
For me it felt obvious that the best way to plan and execute being Club Photographer for a swim meet, optimizing a specific set of goals, was to give Claude a bunch of context, then jam to come up with a plan to optimize the day.
The Context: heat sheets showing who’s swimming when, which side of the pool is easiest to navigate, some ‘banger’ shots I wanted to try for, etc.
The tooling and plan: create a GitHub repo, iterate on a plan in Markdown, and work toward a chronological “game day cheat sheet” on my phone.
It worked really well.
- GitHub repo allowed iteration and refining from my phone while on deck with SLR in other hand
- Markdown file acted as “living document” in my browser
- Claude fearlessly dissected domain-specific formats to extract information I wanted (swimming time standards in a weird binary format)
- All that we learned is memorialized and ready for future swim meets
I know a lot of folks won’t want to use this set of tooling and that’s perfectly fine, and I am sure we will see more seamless integration into other office platforms shortly.
My prediction is that this writeup might age like milk. But in Feb 2026, I found this a concrete and personal example for how once you’ve experienced “Agency” and actual human-agent collaboration, you can optimize just about anything through that lens. Why just ask an LLM questions when you can, with trivially more effort, create scaffolding and a set of scratchpads and work together to optimize or solve a problem?
