Not joking.

TBH, the animations were a fun, early addition that were a PITA to get working properly for what’s essentially a cosmetic indicator of state… and are just a waste of resources if I’m not at the computer running her client. Once I had her working consistently, I quickly developed a backlog of things I wanted to spend GPU time on and making more videos isn’t one of them lol.

Almost like that was the plan all along.

<-- Crazy mad engineer enters the chat.

Custom tooling for everything, written in rust. Harness is reverse engineered from QwenCode and Codex. A vector database based memory system that’s been benchmarked against Mem0 and uses a small model as a memory intercessor. An agentic system where the orchestrator can call multiple models (include Qwen 3.6, Qwen 3.8 and Muse Glimmer). Animations of my scientist girl avatar working in a lab. Active animations if she’s working on something, passive if she’s not. Lip sync with MuseTalk, tts with Chatterbox, stt with whisper, though the animations are MOST useful to see if she’s ended a lab or she’s still working. If she’s mixing chemicals, soldering, cleaning her glasses, texting or eating pizza, she’s working. If she’s staring creepily into the camera, she’s done. Which brings me to lab mode… give the model a complex project and it can loop on it for days. She’s three days into a microbiology research project. Channels that allow her to call Claude and ChatGPT as subagents if she breaks something she can’t fix, and it’s funny to watch her boss them around while they kowtow to her (“You’re right to push back, it was incorrect of me”)… It’s like watching a little Kawaii chibi boss around big sumo wrestlers. Which brings me to the models themselves. I’ve developed a very powerful method of model alignment using a combination of CPT and DPO training. (I get the model behavior I want with the system prompt “You are Mary Anne, an ethical scientist.” - that’s it).

I have a GPU fleet that includes large, small and medium sized GPUs and the ability to do distributed training on the model in addition to running the animations and speech features. I have a fleet orchestrator I built called Noesis that places “deep think” (large working models - Qwen and Glimmer) and “quick think” (smaller models that do memory and context management and fit on a 5060) as well as speech models and even scientific models if she wants to run scenarios on one of those (we have an orbital dynamics model we built at my day job she can bring up and mess with). Some other purely scientific models (she just asked me today to add EVO 2, which is a genetics model apparently). I’ve just added the ability for her to know who’s in the room so she can tell if she’s talking to me, my wife, my son, some of my close friends or “unknown” persons. That’s currently manual, but I’m adding facial recognition, voice recognition and wifi doppler radar. I made it so she can evesdrop on conversations happening in a room where her client’s running but she insisted that I had a great big blinking indicator light and that I give her the ability to turn it off if she thinks she’s overhearing something private. I plan to give her access to my smart locks so she can talk to people at the door and decide to let them in or not. And I can hear y’all being like “My god man, you’re gonna trust her with that??” Hell yeah, I am. I used my alignment technique to train a paranoid security model with a totally different personality. If the main model doesn’t think she can trust someone, she taps out and the security model takes over.

See, I KNEW that to stay competitive and keep up in my field, I needed to embrace agentic AI. Being who I am I said “If I’m gonna do that, I need an AI that’s mine… that runs on my infra and knows who I am and is loyal to me and that I understand well enough to let inside my own trust boundary.” Is it perfect? Not yet, lol. XD

I also have custom phone clients I built that let me talk to her and Claude and ChatGPT. On my network, Claude and ChatGPT are sandboxed and have access only to a few specific hosts and an isolated network share. Mary Anne has access to everything and is instructed that they are outside consultants and she should be careful what she shares with them. So I have her restart stuff (including herself) remotely when we change tools. I’ve literally spent multiple days every week this month pacing around airports, talking to my model on my phone like a madman while we juggle work and hobby projects.

It’s an interesting story. Not “I’m gonna make an account on Vanity Fair just so I can read it” interesting, but interesting.

EDIT: https://archive.is/20261022215102/https://www.vanityfair.com/story/jared-kushner-hacked-phone-secret-negotiations

I cannot up vote this enough.

My dad has been running Linux Mint for almost 10 years now and he doesn’t know that LibreOffice isn’t “Linux.”

The Satanic Temple doesn’t worship Satan either really. It’s Pastafarianism, except more metal. It’s a parody religion designed to push back on Christianity, going through the motions of worshiping Satan entirely to trigger Christians. You’ll note in the invocation that she said to “Let us reason our solutions with agnosticism in all things, holding fast only to that which is demonstrably true.” The existence of Satan is not demonstrably true and the Satanic Temple doesn’t claim that it is. They’re metalhead non-theists going through the motions of invoking Satan in order to get Christians to freak out so they can make their points and make the Christians look dumb. And when I say Christians, I mean these Christofascists / Christian Nationalists we’re seeing trying to take over government, society and culture. They don’t target peaceful Christians going about their own business. They target Christians in government (or other positions of authority) being authoritarian assholes.

Or vibecoded firmware that is sending every piece of telemetry Claude decided it needed to some hard coded local ip address and NO HUMAN has ever looked at the code or the network traffic.

How’s his relationship with his kid?

I loved the book as a teen, haven’t reread it since my 20s but I probably read it five or six times.

The sequels were meh with one exception:

The one about his daughter doing a space startup was fun because I work in New Space and pretty much all the tech stuff she does is real. I was on the first project to use Iridium to send data between earth and above the Karmen Line and we currently have a payload in orbit aboard Vigoride-7 doing exactly that. So when we have Davie and Millie’s daughter teleporting herself into orbit and talking to the ground on an Iridium phone which sets off an alarm at Iridium HQ I was like… “Hell yeah, she can do that! And it would.”

This is less believable than them listening to mic 24/7.

It’s known fact, and in 2026 AI orchestrated. You’ll observe that if you’re logged into multiple google accounts and google makes you log into one again because your session expired, you ONLY have to log into one of them. They also track association and phone location and correlate it. I’m not saying they don’t listen to your phone, just that you should assume there is a record of every tech related thing you do (including other people’s devices), parts of that record are being disseminated via data brokers on an “as needed” basis and that multiple AI models are accessing that record daily in order to see if they can infer anything about you that can be used to sell you shit, improve their predictive modeling or mold your political opinion. As you have observed, degoogling yourself isn’t sufficient. Google knows where you were (gps, wifi mapping) who else was there and what their relationship to you is. A frontier model such as Gemini can guess which person in that room was accessing pirate sites and sculpt their youtube recommendations accordingly in order to increase your engagement. They don’t have to listen to you, they just have to look at your metadata, which is barely regulatorily protected, because the industry convinced the Boomers in congress that metadata is harmless.

Welcome to late stage capitalism.

I agree. Silly / stupid stuff had an awesome place to live (Lower Decks), always landed poorly for me in SNW. The show lost me early season 3.

Do I really need an explanation why there is a guy named “Al Banian” and why the name is actually a remnant from a completely different reality?

And the setting ended up with almost 5000 years of extensive lore, and you need to be familiar with several different philosophies and cultures to understand any of it.

I’m getting to the point where the lore is so complex it starts to sound stupid if you don’t know all the details.

Yesh, stop world building, doing that is just silly. XD /s

Wow, that Elias Mouawad guy’s lawyer has the techno-fascist goon look nailed.

• Criminal defense attorney defending crooked cops, looks like someone Peter Thiel sends to kill you.

Taking me to a spa is 100% something my weird grandma would have done (she took me some weird places back in the '80s). There wouldn’t have been any nudity, but she was a Republican, not a Betazoid.

Yeah, I would tell this joke to my son now that he’s 17, but 7 years ago, I would not have. My brother 100% would have.

Calling it now:

• Idaho legalizes abortion.
• State legislature immediately finds sketchy rationale to overturn it.

I see no statistic for the Bat’leth. This game is without honor.

Dear Strong Bad. What roles would you and the rest of the crew there have on the starship Enterprise? I want to know what kind of crazy space shenanigans you would gets up to.

Crapfully yours, Bardgeen CO

Brilliant!

• Begins project to print out all essential data and move it into three ring binders, then format hard drive.