> Comparing agentic Qwen3.6 35b to Claude Opus is like a junior with knowledge across the board, that you really need to guide, versus a senior that thinks with you on architecture.
that's why i use the frontier models because its a senior co-worker vs a junior. if you use the junior for the sake of privacy i think you're missing out on the best insights for a specific task.
There's a formula I've bookmarked from someone to help you get better social capital. In order to increase your surface for social capital you need to expand your influence, therefore you need to increase your luck:
L (luck) = D (doing) x T (telling)
Basically, you need to be able to show you are able to do stuff (D) and with enough people seeing this by you sharing (T) through different channels, can open doors. I haven't been doing it as heavily but it has already given me incredible opportunities.
i've been thinking about this as well. but having built a startup, i've learned that users don't care as long as they are given the value and most convenience. they don't really care much at security as much as we do. just look at openclaw? but maybe it's our job to make sure it is taken care of vs assuming the user cares and just make it look seamless.
my takeaway from this is that it should now be MANDATORY to have an LLM do a scan on the entire codebase prior to release or artifact creation. do NOT use third party plugins for this. it's so easy to create your own github action to digest the whole codebase and inspect third party code. it costs tokens yes but it's also cached and should be negligible spend for the security it brings.
> The problem for me is that the development practices of the people that are working on it are suboptimal at best; they're constantly releasing at an extremely high cadence, where they don't even spend the time to test or fix things (or even build a proper list of changes for each release), and they add, remove, refine, change, fix, and break features constantly at that accelerated pace.
this is what i notice with openclaw as well. there have been releases where they break production features. unfortunately this is what happens when code becomes a commidity, everyone thinks that shipping fast is the moat but at the expense of suboptimality since they know a fix can be implemented quickly on the next release.
one thing it did massively for me was save me time from questions like should i go with X or Y option questions. before i used to just think longer about tradeoffs but with AI it became a lot faster. no more procrastination due to decision fatigue.
> I simply booted up a VM with an H100, ssh’d into it with Cursor, and prompted the agent to set up an inference server that I could ping from my web generation app. What used to take hours or days of painful, slow debugging now takes literally minutes.
an awesome takeaway from this is that self-hosted models are the future! can't wait for hardware to catch up and we can do much more experiments on our laptops!
i really think this is part of the pitch deck for bun's funding. that a bigger company would acquire it for the technology. the only reason an AI company or any company for that matter would acquire it would be to:
i like how claude code currently does it. it asks permission for every command to be ran before doing so. now having a local model with this behavior will certainly mitigate this behavior. imagine before the AI hits the webhook.site it asks you
AI will visit site webhook.site..... allow this command?
1. Yes
2. No
a concern i have is that it's only a matter of time before a similar attack is done to electron based apps (which also have packages installed using npm). probably worse because it's installed in your computer and can potentially get any information especially given admin privileges.
We wanted to make concept for an app using all local models for chat (llama 3.1 8B) and voice (whisper). Deployed using kubernetes and easily scalable not to mention fully open source!
Came here to say this. What makes a great general purpose language is not only in the programming part but also in transportability. I think Go takes the cake here. If only Elixir apps could be compiled and transported as a binary to similar systems and it works, then it would make it a great general purpose language. Until then, it's really only a good client/server language as it was intended to be.
Such a great quote. Mostly true if viewed especially from a business standpoint. I for one also see code as creative expression, a form of art. I like coding because I can express a solution in a way that is elegant and nice to read for myself and others. A bit shallow but If you've read code that is written elegantly, you'll know that immediately.
Must be my ignorance but everytime I see explainers for LLMs similar to the post, it’s hard to believe that AGI is upon us. It just doesn’t feel that “intelligent” but again might just be my ignorance.
> This weekend I’m working on mobile app that allows you to upload photos and it turns everything into a stitched anime (ghibli or not) with movements and eventually sound and script. Great to make mini animes from your day or travels or anything.
I vibe coded the above including everything: code, design, logos. Just did it solo. It has all error handling, video generation notifications (it takes a while) and credit system. I myself can't believe it's been done in a month with AI. It's already in closed beta in iOS and android app stores. Let me know if you want to try it out before public release.
My quoted comment above was 28 days ago. This is working on this part time and with a family.
This weekend I’m working on mobile app that allows you to upload photos and it turns everything into a stitched anime (ghibli or not) with movements and eventually sound and script. Great to make mini animes from your day or travels or anything.
not to hijack this thread but my dad did extensive research in sea breeze and rainfall modeling and he would have loved to see these AI and machine learning advancements in weather prediction.
that's why i use the frontier models because its a senior co-worker vs a junior. if you use the junior for the sake of privacy i think you're missing out on the best insights for a specific task.