This movement will prove indestructible, no matter how much spin and concealment are deployed to stop it, no matter how many dollars are spent to counter it.
I'm starting to think that, until he's finally gone, I should repost on a daily basis this this 11,000+ word piece I published in @TheAtlantic about @realDonaldTrump's serious psychiatric disorders (Narcissistic Personality Disorder and Antisocial Personality Disorder, for starters) to explain why, as a legal and constitutional matter, he was unfit for the presidency and would only become more so with the passage of time.
In the six-plus years since, every mental health professional I have come across has told me that my essay had it exactly right. I think every word of it has been proven right many times over.
And guess what? He's still only going to get worse.
Gift link:
https://t.co/fPmUjjLcTg
95% of AI engineering is just Context engineering.
Everyone's obsessed with better models while context remains the real bottleneck.
Even the best model in the world will give you garbage if you hand it the wrong information.
Here's what most people miss:
Context engineering isn't just about RAG or memory or agents. It's the art and science of delivering the right information, in the right format, at the right time, to your LLM.
Think about what you actually need:
↳ Retrieval to fetch relevant documents
↳ Short-term memory to track conversations
↳ Long-term memory to remember user preference
↳ Agents to orchestrate everything
↳ Tools to extend capabilities
That's 5 different systems you have to build, connect, and maintain.
I've been building with Pixeltable recently, and it's an interesting approach to this problem. It's open-source and treats context engineering as a unified data problem:
The idea is simple: instead of stitching together a vector database, a SQL database, an embedding service, and an agent framework, everything lives in one system.
Your documents, embeddings, conversation history, and agent outputs are all just tables. Embeddings are computed columns that update automatically. Vector search works alongside your regular data operations.
What I find useful:
↳ RAG pipelines without managing separate databases
↳ Long-term memory through vector search over historical conversations
↳ Multi-agent workflows that persist automatically
↳ Token budget management built into the framework
It's not magical, but it removes a lot of the integration overhead. You're not fighting with three different APIs to make retrieval, memory, and agents work together.
I've shared a starter notebook in the next tweet on building a context engineering pipeline with Pixeltable. It covers all the components and things we've discussed here.
Everything is 100% open-source.
Isn’t it time this was headline news?
#climate
Climate-related disasters forcibly displaced 250 million people globally over the past decade , the equivalent of 70,000 people every day, according to a report by the UN refugee agency
"I wasn’t against communism, but i can’t say i was for it either. At first, i viewed it suspiciously, as some kind of white man’s concoction, until i read works by African revolutionaries and studied the African liberation movements. Revolutionaries in Africa understood that the question of African liberation was not just a question of race, that even if they managed to get rid of the white colonialists, if they didn’t rid themselves of the capitalistic economic structure, the white colonialists would simply be replaced by Black neocolonialists. There was not a single liberation movement in Africa that was not fighting for socialism.
The whole thing boiled down to a simple equation: anything that has any kind of value is made, mined, grown, produced, and processed by working people. So why shouldn’t working people collectively own that wealth? Why shouldn’t working people own and control their own resources? Capitalism meant that rich businessmen owned the wealth, while socialism meant that the people who made the wealth owned it.
I got into heated arguments with sisters or brothers who claimed that the oppression of Black people was only a question of race. I argued that there were Black oppressors as well as white ones. That’s why you’ve got Blacks who support Nixon or Reagan or other conservatives. Black folks with money have always tended to support candidates who they believed would protect their financial interests. As far as i was concerned, it didn’t take too much brains to figure out that Black people are oppressed because of class as well as race, because we are poor and because we are Black. …
[Earlier in my life] When someone asked me what communism was, i opened my mouth to answer, then realized i didn’t have the faintest idea. My image of a communist came from a cartoon. It was a spy with a black trench coat and a black hat pulled down over his face, slinking around corners …
I never forgot that day. We’re taught at such an early age to be against the communists, yet most of us don’t have the faintest idea what communism is. Only a fool lets somebody else tell him who his enemy is… It’s got to be one of the most basic principles of living: always decide who your enemies are for yourself, and never let your enemies choose your enemies for you."
The good thing about the world is you don't have to be clever.
In fact, falling in love with your own cleverness is a major failure-mode.
So clever you fool yourself into saying and doing things that, from the outside, are obviously stupid and selfish and wrong.