codebat
192 posts

codebat
@CodebatG
I do Chip tricks. Founder, Sidecar https://t.co/fjr70dLCN8
Bengaluru, India Katılım Aralık 2022
208 Takip Edilen47 Takipçiler

@SeanZCai Sidecar usesidecar.com is Antikythera for freight forwarding and logistics, and we are seeing intense scaffolding of our users’ long-term tasks so they get outputs and feed the token machine into a better, truer loop that self-improves. Aligned @SeanZCai
English

@jediahkatz `There's a misconception that first-party harnesses from the labs will always outperform` This is also a first party harness over cursor's model: composer.
Not a categorically correct statement
English

Let's talk about why Cursor's agent harness is so good.
There's a misconception that first-party harnesses from the labs will always outperform. For many reasons, that isn't true.
There are roughly 6 layers that go into a good agent harness: orchestration, context, routing, transport, state, and execution (tools). Some of those involve careful context engineering, and others look a lot like the traditional craft of building great software.
Each of these layers needs to be optimized. And if any one of them is degraded, it can severely impact your experience with the agent. This blog post is mostly focused on context and tools, and there's still so much more to talk about there. We also want to spotlight the other (very crucial!) areas soon.
There's a lot more misinformation around the agent harness out there. I'll keep writing about how we build it behind the scenes.

Cursor@cursor_ai
Our agent harness makes models inside Cursor faster, smarter, and more token-efficient. Here's how we test improvements to the harness, monitor and repair degradations, and customize it for different models. cursor.com/blog/continual…
English

@PreslavMihaylov Searching it. Now we are talking good engineering practices
English

@CodebatG there's probably some sort of tool out there which could achieve this using a RAG. perhaps a smarter dependabot with an LLM/RAG integration
English

@HedaSudhanshu @itsmesatwik_ Just beg enough. Gemini is like that kid who is dumb at all exams but is street smart. Useful for legacy vertical AI usecases such as ours.
English





