My AI agents found an unmapped Maya site in Mexico, using LiDAR data from a NASA dataset.
72 hours from a random thought on vacation to a reply from Mexico's national archaeology institute.
What a time to be aliveβ¦
For my agency work I had been designing in the browser for many years. It was the end medium so I thought, why not start there?
I would work with content writers, and wireframe sections. I mostly kept it in black and white so as not to distract with any final polish until the end.
It worked great, and now feels like the way things are done. I know there is still a place to start in design software, but it just feels more and more like that idea of working in the medium things live in is a great direction.
The biggest issue I have with dots is speed and opacity.
Muse is faster for nearly all non-coding tasks (I donβt ask muse to code)
What a dot is doing is opaque. Since Iβm faster to think of things for it to do I can give it a backlog of 4 or 5 things. Then I have to remember what I asked it because of the speed.
the most import skill in 2026 is still being able to tell your agents "wtf is this crap, no, not like this"
if you stop doing that, you're basically fucked, despite what some devfluencers would like you to think.
This is essentially what Iβve been communicating to our team.
Weβre all working on the same block of marble. Engineers are roughing it into form, design is pulling out the intricate tools for the finer details.
@pashmerepat@steipete As I read this list I feel like Iβve already had my dot do all of this on my computers. I noticed Friday it can be connected to more than one at a time but didnβt my dot have ask these capabilities before?
@1Password@bdsams βunless the Safari extension has lost its connection to the 1Password appβ. This happens a lot with chrome on macOS from my experience