@RunicDeveloper
Account based inSouth Asia!
About this account
- Account based in
- South Asia
- Connected via
- South Asia App Store
X says this location may be affected by a proxy or VPN.
Account-level information from X, not a live location or the device used for a specific post.
Building efficient systems and team productivity. Running the runicdevs. https://nitter.cf/t.co/dkEVxdBCyX
Can be found at 👉🏻
Joined April 2025
- Tweets2K
- Following76
- Followers65
- Likes2.1K
New Framer template: Ashgrid 🖤
A dark, cinematic site for AI agencies and automation studios.
→ Original 3D made in Blender
→ Count-up stats, auto-playing process tabs, custom Play cursor
→ CMS case studies + journal
ashgrid.framer.website
Aren't you tired of seeing framer projects looking dead and in 1-3 colours.
We need some life back in the projects and here's the start.
Dinnerbell - a lively and fun template for all to use.
dinnerbell.framer.website
Use the Page Tear Transition component in your next framer project. Customise it as you like and grab now for free for next 7 days.
framer.com/m/PageTear-m2b9LV…
Figma is used by 97% of design system teams, yet 60% have no token automation at all and only 5% have bi-directional sync. The tooling consolidated. The pipeline underneath it never actually got built.
CPython added RISC-V as a tier 3 platform and the IETF standardized the QUERY method so a request can finally carry a body without pretending to be a POST. Boring infrastructure news is the news that ages best.
Anthropic is now used by 43.8% of US businesses and OpenAI by 39.8%, while open-weight models account for 6.4% of spend. The everyone-will-self-host prediction has had two years to arrive. It will first arrive at Enterprise levels, then you will see local compute stations.
You all saw what OpenAI did with the Mathematician’s research. Stole it and named it its own.
Everyone who has been blindly feeding their data and researches into these closed LLMs is asking to be robbed.
The only way forward is fine tuned local LLMs else good old human work.
Use this Password Morph component in your next framer project.
framer.com/marketplace/compo…
Enjoy the new Droplet Reveal framer component from Runicdevs.
Free to use right away!
framer.com/marketplace/compo…
If this incident isn’t reason enough for people to understand that companies like OpenAI and/or Anthropic can steal your inventions and researches if you use their tools, then nothing will be.
Don’t use third party AI for new research / sensitive information.
Launching my new framer component today.
framer.link/dYWwNqP
Check It out. DM for free giveaways.
What the fuck is this @Uber ?
I’m at the latest version of app and still this pop up?
I’m unable to request a ride. Ruddy poor user experience.
The most blatant THEFT
$1 billion to religion centred schools from national tax, not the donations of that particular group is blatant theft and tells you who runs and own the USA, and who has completely invaded and infested it from the inside.
Occupied states of America by Jews
Approximately $700 million to over $1 billion.
New York accounts for the largest share—around $672 million in 2022-23 (Teach Coalition data)—mainly for mandated services, transportation, special education, textbooks, STEM, and security grants benefiting ~180,000 students in Jewish nonpublic schools. Additional funding comes via federal programs (Title I, IDEA) and state vouchers/tax-credit scholarships in places like Florida, Ohio, and Arizona. No comprehensive national total exists; figures exclude pure private tuition/donations.
Debugging tip: read stack traces backwards. Start at the throw site to see the symptom, then move up the call chain to find the intent that broke.
If you care about layout stability, segment Cumulative Layout Shift by component and template. Fix patterns, not pages, and you get more targeted wins.
what the fuck is this @claudeai ?
You are just a fucking tool, not a bloody human, don't lecture me on how to convey commands.
@AnthropicAI this level of censorship/moderation is absolute shit. We are bloody paying you for the tool, not lectures on language.
Before you retrain a model, add a cheap sanity layer that checks entities and basic facts. Catching obvious hallucinations reduces user harm without huge data ops.