Sabitlenmiş Tweet
Pneuma
275 posts
hey @thsottiaux godex is great but sometimes context compaction takes forever(
English
@Korvus_dev good point, my guess - is because of imagegen 2.0 is soo good
English
@scaryinference Game development has a lot of bugs and usually take a long time to make. Codex is much cheaper at building larger projects, building the same project with claude would be way more expensive.
English
nowadays you only read book itself to live through experience described in book. thats where most classic literature lays.
for all other technical/self-help/business books, which are in general just a set of author ideas, u can just feed pdf to codex and ask it create:
- interactive course
- mind map
- game
- whatever format your mind prefer
EP@eptwts
i think nowadays with access to LLM's & deep research agents, reading books is inefficient when it comes to acquiring info... you can get an agent to go through 1000's of books & extract the most applicable info to your specific case without having to waste hours reading if you do it for leisure then that's fine, but reading as a way to "acquire knowledge" is one big cope pushed by the self-improvement cult
English
@_winter_wonders … when there is ratatui and u can make tui on rust
English
can somebody
anybody
tell me what
in the fuck
the point of a TUI written in JavaScript is
English
@rudekyoni999 rpe8 for mostly all sets and never felt extrme muscle soreness
English
@doctormorphh the only supp i really feel - 50g of vodka.
and i dont feel nothing even from 2g of l-theanine, so its fun to read how people “feel extremely focused after 200mg” lol
English
i wish i genuinely felt anything from supplements at all
magnesium, taurine, l-theanine etc i have to take grams of this shit to feel or notice ANYTHING
people swear by noopept or thiamine for photographic memory or that 2g taurine makes them sleepy when i have to do a back flip and land on my neck to feel anything
drMAWZ@TheDrMAWZ
L-theanine is amazing for anxiety My wife loves it - 100-200mg/day
English
AI + ADHD = unfair advantage
88% of neurodivergent employees report being more productive with AI. ADHD curiosity pushes you to explore various things. AI helps you connect the dots, build something unique, and become a generalist who is skilled in different areas of knowledge.
English
BREAKING: Now Claude Opus 4.7 finds you job autonomously! 🤯
Someone build a tool which finds job for you
> Scans job openings at top companies
> Fills out the forms for you automatically
> Rewrites your CV tailored to each position
No recruiter. no sending 200 identical CVs
100% free and open source
English
btw i've made a version which uses Codex under the hood - github.com/6alex6kash6/ca…
just in case you dont wanna be limited by tokens in job searching
ℏεsam@Hesamation
bro created an AI job search system for Claude Code that scored 700+ job applications and actually got him a job. AND IT'S NOW OPEN-SOURCE. It scans multiple company career pages, rewrites your CV per job, and even fills application forms. The repo has: > 14 skill modes (evaluate, scan, PDF, ...) > Go terminal dashboard > ATS-optimized PDF generation via Playwright > 45+ companies pre-configured (Anthropic, OpenAI, ElevenLabs, Stripe...) GitHub: github.com/santifer/caree…
English
@kloss_xyz spot on bortha:
x.com/scaryinference…
Pneuma@scaryinference
you don’t get the 1000 IQ move by anthropic > be a company with not enough compute to satisfy increased users demand > cut limits and make 100$ plan feels like 20$ > users move to codex > good, but load is still high and availability page is red > cut openclaw usage > even more users migrate to codex > now its openai headache to manage mass subsidized subs and increased demand > you can now focus on b2b and deploying mythos to big companies
English
my greatest conspiracy theory is anthropic wants you to cancel your $200/mo claude max plan because they lose money on it… so they make it as painful as humanly possible to weed out the weak… only the survivors who never cancel will receive early access to AGI and the god model
Theo - t3.gg@theo
Claude Code is basically unusable at this point. I give up.
English
@menhguin u just need to fix one thing as adhd:
x.com/scaryinference…
Pneuma@scaryinference
being adhd is unfair in vibeclauding era. only thing u should change: is to switch from doomscrolling to doomterminaling.
English
claude code hitting the ADHD community like crack hitting the ADHD community
Aella@Aella_Girl
claude code feels like it unleashed me. i feel like i was living my entire life with broken legs and suddenly it granted me the ability to run. every project ive ever wistfully thought 'man i wish i had the coding skills to be able to do that' is suddenly in my hands and I'm overflowing. i've never been so creative in my life
English
A great iq test on people who use AI is what level of reasoning they use, if it’s always x-high then they’re probably not that smart
10x'er@10x_er
This stuff is literally not real. I give this feature 6-12 months tops before they remove it, bookmark this and confim. Like what do you mean you want me to pick how hard my model wants to try? I want it to try its hardest, always, and I want it to be fast. Why would I ever need to consider tradeoffs when doing knowledge work with AI.
English
@IanLandsman heres my theory
x.com/scaryinference…
Pneuma@scaryinference
you don’t get the 1000 IQ move by anthropic > be a company with not enough compute to satisfy increased users demand > cut limits and make 100$ plan feels like 20$ > users move to codex > good, but load is still high and availability page is red > cut openclaw usage > even more users migrate to codex > now its openai headache to manage mass subsidized subs and increased demand > you can now focus on b2b and deploying mythos to big companies
English
Hot take. Claude is doing the right thing. Give the real customers a better experience (keep infra working well for them) while ditching people having their bots loop over their calendar 100,000 times a day.
English
@ryancarson the only explanation of this is that anthropic wanna lose users
x.com/scaryinference…
Pneuma@scaryinference
you don’t get the 1000 IQ move by anthropic > be a company with not enough compute to satisfy increased users demand > cut limits and make 100$ plan feels like 20$ > users move to codex > good, but load is still high and availability page is red > cut openclaw usage > even more users migrate to codex > now its openai headache to manage mass subsidized subs and increased demand > you can now focus on b2b and deploying mythos to big companies
English
My whole timeline is everyone talking about the Claude Max plan not being able to be used in OpenClaw.
Will be interesting to see how this shakes out
English
you don’t get the 1000 IQ move by anthropic
> be a company with not enough compute to satisfy increased users demand
> cut limits and make 100$ plan feels like 20$
> users move to codex
> good, but load is still high and availability page is red
> cut openclaw usage
> even more users migrate to codex
> now its openai headache to manage mass subsidized subs and increased demand
> you can now focus on b2b and deploying mythos to big companies
Boris Cherny@bcherny
Starting tomorrow at 12pm PT, Claude subscriptions will no longer cover usage on third-party tools like OpenClaw. You can still use these tools with your Claude login via extra usage bundles (now available at a discount), or with a Claude API key.
English
I think the only way to play the game these days is upskill a lot with AI tools (especially related to domains you care about) then ruthlessly guard unplugged time so you can read & think your own thoughts
People who have spiky brains + machine leverage will dominate everything
English