Seth Juarez

I work in AI/ML @Microsoft / @GitHub. I like machine learning. Oh and the hot Mexican tamarind candy. Don't judge. Married to the beautiful @rivajuarez.

[Seth Juarez @sethjuarez]

The thing AI is actually good for, at least for me: a rough first draft I get to argue with. I do better work when something pushes back.

https://t.co/dwc8HkWEBP

[Seth Juarez @sethjuarez]

@film_girl lies

[Seth Juarez @sethjuarez]

The thesis. A model has no agency. Agency is intent plus action. A model can turn your words into intent, a word or a structure. What it can't do is act. It emits tokens and stops. The doing happens in the runtime.

https://t.co/zEhglXeiRy

[Seth Juarez @sethjuarez]

How LLMs learn, in one breath. Feed it the internet, hide the next word, let it guess, check how wrong it was, nudge it less wrong. Do that a few trillion times. Grammar and reasoning just fall out the side.

https://t.co/5WUtaSV1wh

[Seth Juarez @sethjuarez]

The Luddites weren't anti-technology. It started as a labor movement, workers defending their trade, serious enough that Lord Byron spoke for them in Parliament. We remember it wrong.

https://t.co/dwc8HkWEBP

[Seth Juarez @sethjuarez]

How does a next-token guesser write paragraphs? A loop. Guess a token, glue it on, feed it back, guess again, until it hits a "done" token. That's autoregression, the engine under every chatbot you've used.

https://t.co/5WUtaSV1wh

[Seth Juarez @sethjuarez]

Post two is up. The model doesn't have agency. It just emits tokens. The agent does, because the agent is the model plus a runtime. Watch the handoff happen, one shape at a time.

https://t.co/zEhglXeiRy

[Seth Juarez @sethjuarez]

"Just learn a new skill" is what we tell people about AI now. Someone handed wool workers the same line in 1786, as machines took their jobs. They weren't impressed either.

https://t.co/dwc8HkWEBP

[Seth Juarez @sethjuarez]

A language model does exactly one thing. It guesses the next token. Not the answer, the next token, then it stops. Everything we call "agentic" is scaffolding around that one move. Still amazes me it works.

https://t.co/5WUtaSV1wh

[Seth Juarez @sethjuarez]

Found a petition from 1786 begging Parliament to ban a machine for killing jobs. Swap in "AI" and it's eerily close to today's headlines. We've been doing this for 240 years.

https://t.co/dwc8HkWEBP

[Seth Juarez @sethjuarez]

New series on how AI agents actually work, starting from the plumbing instead of the demos. Post one climbs all the way down to the one dumb thing a model actually does. It guesses the next word.

https://t.co/5WUtaSV1wh

[Seth Juarez @sethjuarez]

The first industrial revolution made muscle cheap. This one is making thinking cheap. Still not convinced that's the bad news everyone assumes it is.

https://t.co/dwc8HkWEBP

[Seth Juarez @sethjuarez]

I haven't written anything in about six years, so naturally I came back with a rant about 240 years of people trying to ban machines. anyway, new post.

https://t.co/dwc8HkWEBP

[GitHub @github]

Cooking up something new 🧑‍🍳

Join the waitlist for early access to technical preview of the GitHub Copilot app 👇
https://t.co/ODODKdvzOA

[Satya Nadella @satyanadella]

2/ And with our Rubber Duck agent you get a multi-model reflection loop as a reviewer. GPT-5.5 can review the output of another model or vice versa!

[Satya Nadella @satyanadella]

Super excited GPT-5.5 is rolling out to GitHub Copilot, M365 Copilot, Copilot Studio, and Foundry today.

With deeper reasoning, stronger multistep execution, and better performance across long, complex tasks, GPT-5.5 helps you go from idea to execution faster with fewer iterations to get to the right outcome.

It’s all about helping you choose the right model, or models, for the right task across your workflow.

[Seth Juarez @sethjuarez]

@quorralyne OK - made a fix that makes it more obvious

[Seth Juarez @sethjuarez]

In an effort to start writing things down again, I took a baby step and wrote down a bunch of ideas of things I might turn into posts. What should I write about next? https://t.co/0CWkCaB720

[Seth Juarez @sethjuarez]

@quorralyne I think I need to redo some UX - it lets you vote if you're signed in (there's no password to make or anything). Will make some adjustments.

[Aspire 💫 @aspiredotdev]

HAPPY NEW YEAR! #AspiriFridays is starting back up TOMORROW - finishing what we started in 2025 with @sethjuarez! See you at 12pEST/9aPST! https://t.co/5v7iJdp2CW