@cline 1. Multi-Agent Orchestration The sci-fi vision of agents (‘rear agent, quarter agent, analyzer agent, orchestrator agent’) sending out a swarm of sub-agents and combining their results sounds cool but in reality, most useful agentic work is single-threaded.

@cline Complex orchestrations rarely deliver real value and usually add confusion. Its hard enough to make the models work in a single thread let alone doing all this parallel orchestration logic that adds not just implementation complexity but model interpretation complexity.

@cline 2. RAG (Retrieval Augmented Generation) for Agents RAG is a mind virus. It seems powerful on paper, but in practice even something as basic as GREP often works better, especially for agents.

@cline The RAG hype doesn’t translate to practical agent workflows because RAG leads to really scattered code that doesn't lead to a useful "understanding" for the model.

@cline Its almost always better for models to list the files and search through them with grep and then open and read the whole thing(Like a person). @cline was using this since forever but @ampcode and @cursor both shifted towards it.

@cline @AmpCode @Cursor 3. More Instructions = Better Results The myth that stacking the system prompt with more and more “instructions” leads to smarter models is flat out wrong. Overloading the prompt confuses the model as more instructions often lead to conflicting advice and sensory overload.

@cline @AmpCode @Cursor You end up playing whack-a-mole with behaviors instead of getting useful output. For most frontier models today its just better to get out of the way of the models rather than the constant yelling at them to steer them a certain way. Measure your words(tokens) carefully.

@cline @AmpCode @Cursor Once again, all these ideas are very enticing and if you didn't spend all day tinkering with AI you would think they all make sense except that they don't. Our stances will them will change as underlying models improve.

@arafatkatze @cline GREP is still RAG thank you for your attention in this matter

@jeffreyhuber @cline GREP is RAG but its not Vector DB which is what I meant(There's a photo too) and every normal person reading this understands that. Are we really playing semantics here?

@arafatkatze @cline boomer take tbh

@TheAiComposer @cline Okay zoomer, what did get done this week?

@arafatkatze @cline Now tell us about the agent framework mind virus, since Cline doesn’t use one.

@arafatkatze @cline What?

@arafatkatze @cline Exposing the AI snake oil!

@arafatkatze @cline

@arafatkatze @cline Curious if you’ve read this post about how Anthropic uses multi-agents for their Research feature? The results seem to apply to more research based tasks but the results (90% improvement with multi-agents over just Opus) and blog are super interesting! https://www.anthropic.com/engi...

@arafatkatze @cline How do you compare with the experience of @AnthropicAI with multi-agent systems - https://www.anthropic.com/engi...

@arafatkatze @cline Thank you for putting this out there

@arafatkatze @cline Agreed, complexity for its own sake often ruins clarity. Simplicity and focus usually win.

@arafatkatze @cline Stop with this mind virus nonsense