So, let's see what we've learned so far.

"A.I." coding assistants produce better results ("value earned/tokens burned") when you:

* prompt with tests
* work in small feedback cycles with continuous testing, code review, refactoring and integration
* cleanly separate concerns in your code

LOL

Spending some time doing a little art and design for the first time in years.
I’d forgotten how awesome it is to try a couple of cheap experiments and stumble on a new effect that really slaps.

This is what iteration should feel like.

Went out walking a couple of weeks ago and found this character.
Complete menace, assaulted another squirrel and nicked their food!

Very photogenic though.

#squirrel #wildlife #nature #dusk #photography

just trust me and read this

Wallabys.

A fun heuristic I've adopted when coding. (If you're into that sort of thing.)

Leave the tests red.

1/
I used to make a point of "finishing" a commit/ticket/test/whatever before stopping work. More recently, I try to leave something *unfinished* before stopping. (Ideally with a red test!)

🧵

Do you like stories? I have a story.
1/
As the oldest grad student in my lab a long time ago, I was also the default system administrator. None of us knew anything about system administration, but I was the lucky one tasked with sorting it out.

🧵

I've heard from colleagues that adding abstractions in code makes their work too complicated. I want them to be empowered to make design decisions, but not have DTOs in the business logic.

I find this stuff fun. I'd love to try and infect them with that enthusiasm, but I'm not sure how...

"based on what you know about me. draw a picture of what you think my current life looks like"

For the sake of my neighbors, lets hope not!

Had a lovely time yesterday at readingtechfolk.org

Engaging conversations on topics like automated architecture testing, organizing Terraform in projects, and the future of open source.