Modern aviation is an amazing engineering feat (all the environmental issues aside). With such numbers it means that "one in a million" failure occurs roughly evey week, but the fact that we have so little major incidents demonstrates how well aicrafts are engineered.
Excuse my likely stupid question, but has anybody had some success using Claude Code with frontier agents (or Junie or anything else) to invoke local LLMs for specific sub-tasks or wrapped as skills? In other words, is there a way to use expensive, frontier models as orchestrators that manage local models to do the specialised coding tasks?
That is my entire workflow in opencode through delegates. I have an "orchestrator" which uses matt pocock skills like grilling and writes specs in gh tickets. Once that is done it delegates individual tickets to deepseek v4 flash based developer subagent which executes the code pretty fast. This agent can only delegate one subagent which is a code-reviewer which uses a better model like qwen 3.7 for review. So dev does its own review loop before returning.
It's going amazingly. The orchestrator holds the big knowledge from grilling and also enables me to do more grilling to refine the specs. The trick was to check and re-align after each phase was developed. Also carefully defining each workflow step explicitly otherwise it makes mistakes like trying to self review the code. Also needed to define very explicit contracts. I chose the same surfaces as the matt skill set.
The delegate builder is too ugly to share yet. But to be honest just use the normal builder with a frontier model and instruct it to delegate to developer any developer task. That works well.
The system as a whole is meant to support that use case, where each task (ticket in its jargon) can be tackled using a custom workflow that can each use a different agent/llm (so, it should support local LLMs if you have configured your coding agent to use them).
Sidenote: it's still not where I want, but getting there...
This may not be exactly what you're asking, but I've been running Hermes using different frontier models (like DeepseekV4-Pro, GLM-5.2, and others) as the heavy lifters but who also spawn agents to run my local models (mainly Qwen3.6-IQ4) for offloading tasks that are a good fit for a quantized model that's local.
Works really well.
How do you know that they don't train their models or append your prompts to add backdoors, or compromise your supply chain by including evil dependencies? This seems hugely irresponsible.
> How do you know that they don't train their models or append your prompts to add backdoors, or compromise your supply chain by including evil dependencies?
I think most of these discussions aren't about irresponsible vibe-coders, as that whole thing is mostly a fun joke more than something serious. The rest of developers who use LLMs for development, review the code the agent writes, iterates and makes changes. Think more like pair-programming, than "Write me X then deploy to production".
I know Twitter makes it seem like everyone is doing vibe-coding and YOLOing podman images into production, but it's very uncommon in a serious/production environment to act like that. While a proper structure doesn't make it impossible for the LLMs to add backdoors either via dependencies or otherwise, but it sure makes it a lot harder.
Personally, LLMs are barely able to work alongside developers and not miss anything, I wouldn't be so worried about them being able to do normal work + malicious work at the same time, as they barely handle the first part properly yet.
> How do you know that they don't train their models or append your prompts to add backdoors, or compromise your supply chain by including evil dependencies?
That's a nice theory, but it doesn't work like that in most EU countries where banks are reluctant to give you a loan even for the 2nd property, let alone for tens of it unless you have collaterals to cover the cost.
Of course. Both desktop and mobile version as well with uBlock Origin. Don't have any issues so far except Google Meet not being able to blur the background on the video calls, but I can live with that.
I don't advocate or support such malware instalation. Just saying that "Russian citizens have nothing to do with this war" is false. There are better ways to express your support for Ukraine, for example, make malware targeting only military or govermental institutions that are directly involved.
This sort of logic justifies terrorist's killing of civilians like 9/11. If they can quote your book, the US citizens voted the governments in that oppressed and bombed members of their religion.
Majority of average American citizens support their government's actions against Iraq/Syria/Libya/Vietnam/... . So in my book, they are also responsible.
It may be the case but no one knows for sure. There is no independent sociology in Russia and the state controlled one is a gear in propaganda machine. Also many Russians are afraid to speak what they think and when asked just repeat what they heard on TV even if they don't agree.
yes, but the majority of russian citizens don't install node modules. This is hitting a demographics where you have the least possible support for war and regime (which is also not small, but maybe, hopefully, not a majority)
It was proven over and over again, that the more you give to Russia, the more they will take next time. We have to stop them now or never. War is already on going between Russia and the west (Putin himself and his chief of spies already said, that they deem sanctions akin declaration of war), hiding from this fact won't help.
Why Telegram? I see Ukrainians use it a lot in this conflict,but I would really not trust it to be secure against such powerful adversary as Russia. Signal is much better bet under these circumstances.
