moon of hearts

Zero code. Just AI. 🤖 Building an EdTech app for 1.5M+ students taking the world's most competitive exam. Follow the build. ⚡️

[moon of hearts @Arkifeel]

@maxzrco i'm at zero. so, yeah Hi! buddy

[moon of hearts @Arkifeel]

@piyushgaarg interested ☺️

[moon of hearts @Arkifeel]

fuck ,antropic won again on even geting banned first
openai losess on this also

[Tibo @thsottiaux]

New moon. New models. Welcome GPT-5.6 Sol, currently in limited preview.

[moon of hearts @Arkifeel]

@anshuc can you write article about this ? very helpful , i got kinda confused as how to maintain the flow and all

[Anjali @anjalid2006]

Any Marvel fans who are waiting for spiderman brand new day???

[moon of hearts @Arkifeel]

@yapper_so earn

[moon of hearts @Arkifeel]

@original_ngv @iitm_bs congrats

[moon of hearts @Arkifeel]

@RoundtableSpace this https://t.co/lxUJCx4gei

[moon of hearts @Arkifeel]

my take - Time will tell !!

[Lisan al Gaib @scaling01]

Cold take on what comes next:
- OpenAI will flourish
- Anthropic will continue to be profitable
- Google will not catch up to Anthropic or OpenAI
- no chinese company will catch up to Anthropic or OpenAI
- the highest tier of intelligence will become a luxury product that only companies and multi-millionaires/billionaires can afford
- most of the companies that invested massively in them will have massive returns
- SpaceX’s AI will be fine and on par with Google by end of year
- Nvidia will become the first 10T company

[moon of hearts @Arkifeel]

Here it is , guys no sonnet !

[Claude @claudeai]

Introducing Claude Opus 4.8: it builds on Opus 4.7 with sharper judgment, more honesty about its own progress, and the ability to work independently for longer than its predecessors.

Available today at the same price.

[klöss @kloss_xyz]

Most people make average skills with Codex, Claude, Cursor, OpenClaw, Hermes, and Grok.

How to make yours production grade:

1. Define the goal in one paragraph. What inputs, what outputs, and what failure modes.

2. Send the AI to deep research public GitHubs, existing skills, and the workflows of people already shipping in that space.

3. Have it turn the research into a plan with named sections, examples, and references.

4. Stress test the plan. Ask what’s missing, what’s wrongly assumed, and what breaks at scale. Make it test its own hypotheses against the references it pulled.

5. Package the skill in full. Install it and run it on multiple real tasks end to end.

6. Feed the failures back to the model. Tweak, iterate, and finalize the package.

7. Implement Karpathy’s autoresearch.

Now you have self-improving production grade skills you can use universally.

Anyone can make a skill now.

Your loop is where you stand out.

[moon of hearts @Arkifeel]

openAI i hope they did great over this one

[moon of hearts @Arkifeel]

Hey @antigravity , so your CLI can't even register a custom slash command from a workspace folder? Seriously? Claude Code figured this out on day one. I literally have SKILL.md files sitting right there and your CLI just goes "unknown command" like it's never seen a .md file .

[moon of hearts @Arkifeel]

google Fucked openclaw!!

[moon of hearts @Arkifeel]

Sometimes X recommendation is top tier: https://t.co/XjrfjACTrO

[moon of hearts @Arkifeel]

@reyTempest_ Blue lock ,rent a girlfriend.