Andrej Karpathy: paying $20/month for a subscription isn't "using AI" - most people never even started
the gap was never access. everyone has the same model, in the same tab, one click away.
the dead list: copy-pasting prompts all day, arguing with the model in circles, rewriting the same code 5 times, asking it questions like it's google, "the AI just doesn't get it"
what actually separates the top 1%:
→ they configure the tool instead of just opening it
→ they give it context, rules and memory, not one-off prompts
→ they build systems the model runs inside, not single chats
→ they think in workflows, not in questions
→ they make AI do the work while everyone else watches it type
that's the real skill. not prompting harder. building around it.
most people sit next to the tool. almost nobody makes it work for them.
one group types questions. the other group ships systems.
you're not behind on access. you're behind on leverage.
how to build anything rn:
- get a hetzner, do, or hostinger vps
- host hermes on it
- add gbrain or implement your own memory vault using qmd + sql
- set up hermes with codex auth -> gpt-5.5 / no reasoning / fast mode
- install orca on your macbook and phone with tailscale to have a nice ide to work on both
- before starting any work, ask hermes to conduct deep research on the subject and save it to gbrain as source material for the project
- use the `/grill-me` skill or a similar prompt to uncover as many unknowns as possible. save results to memory too
- define/write clear evals for every project to determine whether a run was successful
- have hermes iterate over the project until all evals pass, saving all learnings to the vault along the way
- whenever it gets stuck, use memory + a new research or `/grill-me` session to unblock it
rinse and repeat until the work is done. pay attention to the process. develop a feeling for how long tasks should take and do not be afraid to stop a model mid session to ask for status and why it's taking so long.
@beffjezos@elonmusk@iScienceLuvr Finishing PhD is a very strong signal for self-discipline, resilience and mental toughness though in an academic world where grad students' mental health is at the lowest..
“But technology is the real skin of our species. Humanity, correctly seen in the context of the last five hundred years, is an extruder of technological material. We take in matter that has a low degree of organization; we put it through mental filters, and we extrude jewelry, gospels, space shuttles. This is what we do. We are like coral animals embedded in a technological reef of extruded psychic objects. All our tool making implies our belief in an ultimate tool. That tool is the flying saucer, or the soul, exteriorized in three-dimensional space.”
― Terence McKenna
Personal update: I've joined Anthropic. I think the next few years at the frontier of LLMs will be especially formative. I am very excited to join the team here and get back to R&D. I remain deeply passionate about education and plan to resume my work on it in time.
Auditionnée devant la Mission d'information sur l'IA, la neuroscientifique Ghislaine Dehaene-Lambertz (CNRS) démonte "l'étude du MIT" qui avait buzzé l'été dernier (voir ci-dessous) et selon lequel ChatGPT diminuerait l'activité cérébrale et atrophierait nos capacités cognitives.
was in line to get into one of the biggest and oldest librarys in the heart of Paris, then I, among other students/people in the line, get approached by a group to talk about dangers on AI and the industrialisation around it.
oh man they heard the e/acc inside me debate hard until it was my turn to get into the library..
i launch a script over ssh in my remote VM from my laptop on a live terminal, laptop is closed -> looses ssh connection -> script running gone. with tmux i can just lose the ssh connection without worries, so in a sense it is tmux because of the way we all use ssh for dev on remote VMs
You know what, yeah. I feel silly for not trying earlier
we should just rewrite everything Python/js in Rust now
Python was the language for chat assistants
Rust is the language for tool use agents
It's harder for humans, but I see how it noticeably raises the IQ of LLMs
It’s 2018 and your coworker just sent you a 400 line pull request.
You get a cup of coffee and sit down to review it.
It’s beautiful. Elegant micro-refactors. Crispy method names.
You catch a few things, but that’s ok. It’s part of the dance. They didn’t consider extensibility on part of their API. Here’s a comment buddy.
They respond in an hour saying they think we should do one piece differently than your comment. Hey let’s jump into a room and figure it out. We can’t just agree to disagree, this code is too important.
The PR merges and goes to prod. You feel a shared sense of ownership and accomplishment.
That night you go to sleep and dream of that code. You can still see the shapes of it on the backs of your eyelids, your IDE syntax highlighting sparking neurons in your reptile brain.
You go to work the next day ready to go. You understand the system. N is your foundation. Time to build n+1.