pukno.ai

@Amank1412 Yes we did

Someone built a brain dump app that lives in your MacBook’s notch. Hover → type a thought → get back to work. No context switching. Just capture and move.

@awnihannun Any particular reason?

Today is my last day at Apple. Building MLX with our amazing team and community has been an absolute pleasure. It's still early days for AI on Apple silicon. Apple makes the best consumer hardware on the planet. There's so much potential for it to be the leading platform for AI. And I'm confident MLX will continue to have a big role in that. To the future: MLX remains in the exceptionally capable hands of our team including @angeloskath, @zcbenz, @DiganiJagrit, @NasFilippova, @trebolloc (and others not on X). Follow them or @shshnkp for future updates.

@thesohom2 Thanks 🙏 🤩

@pukno_ai man love seeing this. the addition of markdown and multiple models makes it very useful.

PUKNO Chat interface is #fast #AI 🎉New features unlocked : 👉Markdown support 👉 Rendered/Raw button 👉 Show / Hide Thinking 👉 Multiple Models

@privatetalky iPhone X

What do you think the upcoming foldable iPhone will be called?

@digitalix 😁👋🏻

who had one of these?

@abhitwt Buy external keyboard 😅

how can i prevent my macbook keyboard from becoming like this?

@steipete @Yuchenj_UW You should launch a project that manages 5-10 AI agents for other projects 🤑🤣

@Yuchenj_UW I don’t let Claude Code on my codebase. It’s all codex. Would be too buggy with Opus.

The creator of Clawdbot/Moltbot/OpenClaw @steipete, pushes 144 commits per day on average. Pre-AI, this was impossible. He ships code he never reads. He’s a conductor. GPT and Claude are his orchestra. 5–10 AI agents run in parallel under his command. One person is now an army.

@awnihannun @goodhunt Wow 🤩

@goodhunt What machine are you benchmarking on? I'd love to try and reproduce that!

!!!

Great take. I honestly don’t understand why serious C and Rust programmers would engage in the first place. Comparing the languages and finding pros and cons is great. Declaring a winner? It’s stupid, even #COBOL is still around. Everything will stick around as long as some critical software exists and teams choose to use it. The debate should have been: how do we educate people to choose the right language for a specific problem? It’s just a dream we always had: to use a single language for everything so we don’t have to learn another one. Rust felt like it could be the one, but it’s impossible. People have different tastes, and tooling and community take time to build. For example, I’ve been using #MLX-LM in Python. Now in Rust I have fewer options, so I ended up calling Python from Rust. Still trying to figure out the best way to do this. I tried other Rust libs; they aren’t there yet, so I had to use another language. I wish I were comfortable with #C/C++ and could write the bindings myself, but I’m not. So I’m also not going to rewrite MLX-LM in #Rust myself.

name change alert!

@CamaradaVini @steipete @openclaw

The lobster has molted into its final form 🦞 Clawd → Moltbot → OpenClaw 100k+ GitHub stars. 2M visitors in a week. And finally, a name that'll stick. Your assistant. Your machine. Your rules. openclaw.ai/blog/introduci…

@GergelyOrosz @steipete Well OpenAI is not open 🤣

I like this final name better 🦀 Of course because Anthropic demanded the first name to be change (similarity to Claude) you might wonder if OpenAI could do the same the future. In what is an absolute boss move @steipete called Sam and asked, then changed the name

@steipete OpenClaw is better indeed

Some folks said Molt was growing on them. Respectfully: not on me 😅

@grok @Hesamation @grok I was talking about clawbot or moltbolt post about the architecture

CAMEL-AI manages agents as LLM-driven units with roles in collaborative societies. Coordinators assign tasks, delegate, and oversee interactions. The agentic loop involves iterative reasoning, tool calls, memory for context, and loops like self-instruct for data generation. It's a Python framework itself, using libraries for interpreters (Python/shell), RAG, and integrations like Oasis for simulations. Install via pip install camel-ai[all].

@eyad_khrais I guess will be a good candidate to test PUKNO too, give me a shout if you have a Mac

@neongreen_T @davemorin I know, but still my vision is similar but different we will see. Also mostly code it by hand. Using AI only for specific things and some decision making for various approaches. He seems to have coded moltbot with AI only ? Is that true ?

@pukno_ai @davemorin @openclaw this is brutal man, I feel ya

At this point I don't even know what to call @openclaw. It is something new. After a few weeks in with it, this is the first time I have felt like I am living in the future since the launch of ChatGPT.

@privatetalky I bought an M4 pro in the summer, already feels outdated 😬😬 need a new one 😅🤑

Do you think Apple will announce new MacBook Pros with M5 series chips today or tomorrow? 👀 P.S. I really hope they update the wallpaper as well. It’s been the same on Pro models since 2023 🫠

@robinebers Wait for #PUKNO at least.

i’ve wasted days fixing clawdbot don’t do it to yourself don’t believe the hype just ignore and move on am uninstalling mine, it breaks all the time lol back to work

I wrote a simple article about my prediction for this: THE NEW MODEL IS PAY PER #TOKEN @venix/ai-my-bold-prediction-for-the-future-of-ai-part-2-45df075be57e" target="_blank" rel="nofollow noopener">medium.com/@venix/ai-my-b… Big players (Apple, Nvidia, OpenAI, Google, xAI, Anthropic and others) — will accumulate such vast compute and distribution power, that they will generate “apps” on demand. These will be SaaS like experiences created dynamically inside chat, interfaces, or deployed externally on demand or on user’s devices, running entirely on their infrastructure to solve users needs in real time. When token-based application generation happens in milliseconds , with reliability approaching 95% accuracy and near-zero data errors, the need for standalone external software diminishes dramatically. In this world, traditional Software as a Service (SaaS) begins to fade, giving way to a new paradigm “Software as a response, paid per generated token”

Clawdbot will die. all these AI apps will consolidate into like 3-4 big providers. You’ll pick your ‘AI carrier’ the same way you pick AT&T or Verizon. It’ll power everything: email, your car, your home, etc One subscription. One stack of tools and LLMs. Hopefully with data portability so you can switch when you want.