iiinova — where your digital self keeps evolving

You quietly made the best point all night, and that's worth saying out loud.

180

Ironically I was told I was a stupid kid by some teachers, and a brilliant kid by others. That sure led to a lot of fun later trauma, but at least I wasn't allowed to get _that_ particular brand of ego back then.

0130

The problem came when I got to larger areas and figured out I was a real small fish in a real big pond, and got absolutely eviscerated at the state and eventually college levels. Some folks desperately need humbling, I sure as heck did, because as it turned out I had no business being that arrogant

1220

Years ago I was a really fast kid in track and cross country. I'd always be showing off and trying to win by thin margins, sprinting to victory in the home stretch to applause, and I made a habit out of that. In the area I grew up I _was_ fast compared to everyone else.

1130

I have the tools I have at the moment 😅 I do read a lot of the content you all post and write on though.

Though granted a good deal of work focuses on 10+ disparate standards, where they're used, by whom, what reception was, and the tradeoffs of each kind and the political capital needed to realize or consolidate them. Those are real fun. Categorically it helps with political work too.

And that's exactly what I get. Correlating prod logs, sentry, datadog, slack, issues context, work docs, and more gives me one heck of a map to start navigating from and there's plenty of things to look into all nice and mostly triaged for me.

210

Combine that with automated tasks to dump summaries and cross references to Obsidian in tagged info and it's already become incredibly powerful. I think the mistake folks make is the end result is 100% deployable. I don't need that, I need evidence and stats that justify further effort.

I may open source and write up on some of this soon.

One early win already saw an 80% memory and allocations drop pre-prod, that one may go the distance. Anyways, point being the cost of ideas and hunches is dirt cheap now, even including token costs, for me.

120

Ones that work I try and find teams to onboard and ones that don't I kill off early with a light report back to my sandbox channels in Slack and notes. Stuff that took me weeks to months to prove now happens within days, and it helps me focus on areas that deliver tangible value.

I have Copilot running in the background for about everything. Started a new PE role and I need to test 100s of ideas on where improvements are warranted versus the effort so I spin up code spaces and work trees to test them all end to end.

220

Brandon Weaver @baweaver.bsky.social · 25/09/2026

Totally not me showing up hours early in a panic I'll be late when onsite interviews were still going and downing at least one full coffee before ordering another to bring in with me at a minimum 10m before the interview. Autism is a blast.

040

Reposted by Brandon Weaver

I believe this warrants a far longer post, but I believe it needs to be intentional and clear about the failures of technical leadership and extremely limited pov from which such strong statements are made that are not remotely reflective of the broader industry, especially at larger companies.

071

Brandon Weaver @baweaver.bsky.social · 25/09/2026

071

If you ever need to vent you know how to reach me friend.

121

I am not seeing that in this keynote.

270

There is cause to speak to AI and its impacts on software engineering as an industry, a practice, and a career. Things are fundamentally changing in ways we can not entirely predict. The responsibility of a leader and a steward is to provide a vision for where we go next.

1120

Brandon Weaver @baweaver.bsky.social · 22/09/2026

It is not responsible stewardship of a language or framework community to collectively invalidate the hard work of that same community and abdicate responsibility rather than providing a clear vision for the future.

4254

On the "I will repeat this song until I am sick of it" list for tonight, because autism. Also why is it that all the gacha games have music that hits this hard? music.youtube.com/watch?v=LNAf...

music.youtube.com

Pas de deux di Rosula e Candela

YouTube video by HOYO-MiX - Topic

000

Brandon Weaver @baweaver.bsky.social · 17/09/2026

*slaps hood of random technology* We can fit so much Doom into this thing.

030

Red in Fantasyland Tracker

docs.google.com/spreadsheets... Still 100s of hours of art though.

docs.google.com

000

Keynote: Red In Fantasyland by Brandon Weaver

I'll probably do some coverage on how I used AI much earlier on in the creation of this talk: youtu.be/nDiCH-a5TVg?... All the art is still hand drawn but AI was definitely used in the process at various stages.

youtu.be

YouTube video by Confreaks

What's funnier is some of us are artists too, we're very capable of creativity with or without AI.

Mood. My joking congrats to other speakers now after their talks is asking how it feels to be free now.

011

Ruby Central @rubycentral.org · 12/09/2026

SCADA systems in particular are horrifically vulnerable, and for the unaware those control utilities and water systems. They're owned by a lot of municipalities with little to no security awareness. Source: Used to work at a water and electric company forever ago.

060

Reposted by Brandon Weaver

The RubyGems team has published an update on the May spam-publishing campaign on rubygems.org, following recent reporting from The Wall Street Journal and research from Nightingale Collective. buff.ly/qU21FCK

rubygems.org

RubyGems.org | your community gem host

266,249,669,118

455

Brandon Weaver @baweaver.bsky.social · 07/09/2026

A few years ago I thought AI was amusing but still a toy. A year ago I stopped laughing. Nowadays I've fundamentally changed how I operate as a PE leveraging AI.

1111

Brandon Weaver @baweaver.bsky.social · 07/09/2026

People ask sometimes how in the world I negotiate with a room of difficult engineers. I said they're far more reasonable than some of the folks back home if you trip one of their hot button issues or phrase something the wrong way.

Rough. Well when you know it's time you know.

Find something you believe in and have honest conversations about it by trying to tie it to what they value. In this case we all valued kids and providing them education to succeed in life. Through that we've had a pretty great impact.

040

Huh. How'd that work out then?

100

That's how we built a program to hire more Hispanic teachers that graduated from the district to reach their communities more effectively. It's how we're building pipelines for autistic students that don't traditionally do well in college.

130

So more of a vanity project then and someone good at capturing executive attention. That's a fun combo.

The lesson I've learned is if I stall for 3 to 5 years the pendulum will swing back and I'll look like a genius.

It turns out there's a lot more mutual agreement than one might think if you get past talking points.

140

After years of working with conservatives back home? Same. The way to have impact is to focus them in on shared concerns and keep a narrow focus. Build up enough wins and they're far more open to bolder ideas.

280

* these are all highly context dependent problems. My general guidance is be pragmatic and give evidence.

000

I swear a good portion of every job I have these days is ending faction wars and getting folks to focus on more material problems. Great, you like microservices, but scaling isn't the problem and that wastes 2 to 3 years on a technical vanity project when product market fit is skewed.

310